1152 - occupational safety and health administration

fede

ral r

egiste

r

1151

ThursdayJanuary 8, 1998

Part II

Department of LaborOccupational Safety and HealthAdministration

29 CFR Parts 1910 and 1926Respiratory Protection; Final Rule

1152 Federal Register / Vol. 63, No. 5 / Thursday, January 8, 1998 / Rules and Regulations

DEPARTMENT OF LABOR

Occupational Safety and HealthAdministration

29 CFR Parts 1910 and 1926

[Docket No. H–049]

RIN 1218–AA05

Respiratory Protection

AGENCY: Occupational Safety and HealthAdministration (OSHA), Department ofLabor.ACTION: Final rule; Request for commenton paperwork requirements.

SUMMARY: This final standard, whichreplaces the respiratory protectionstandards adopted by OSHA in 1971 (29CFR 1910.134 and 29 CFR 1926.103),applies to general industry,construction, shipyard, longshoring, andmarine terminal workplaces. Thestandard requires employers to establishor maintain a respiratory protectionprogram to protect their respirator-wearing employees. The standardcontains requirements for programadministration; worksite-specificprocedures; respirator selection;employee training; fit testing; medicalevaluation; respirator use; respiratorcleaning, maintenance, and repair; andother provisions. The final standard alsosimplifies respirator requirements foremployers by deleting respiratoryprovisions in other OSHA healthstandards that duplicate those in thefinal standard and revising otherrespirator-related provisions to makethem consistent. In addition, thestandard addresses the use of respiratorsin Immediately Dangerous to Life orHealth (IDLH) atmospheres, includinginterior structural firefighting. Duringinterior structural firefighting (an IDLHatmosphere by definition), self-contained breathing apparatus isrequired, and two firefighters must beon standby to provide assistance orperform rescue when two firefighters areinside the burning building.

Based on the record in thisrulemaking and the Agency’s ownexperience in enforcing its priorrespiratory protection standards, OSHAhas concluded that compliance with thefinal rule will assist employers inprotecting the health of employeesexposed in the course of their work toairborne contaminants, physicalhazards, and biological agents, and thatthe standard is therefore necessary andappropriate. The final respiratoryprotection standard covers an estimated5 million respirator wearers working inan estimated 1.3 million workplaces in

the covered sectors. OSHA’s benefitsanalysis predicts that the standard willprevent many deaths and illnessesamong respirator-wearing employeesevery year by protecting them fromexposure to acute and chronic healthhazards. OSHA estimates thatcompliance with this standard will averthundreds of deaths and thousands ofillnesses annually. The annual costs ofthe standard are estimated to be $111million, or an average of $22 percovered employee per year.

DATES: The final rule becomes effectiveApril 8, 1998.

Compliance: Start-up dates forspecific provisions are set forth in§ 1910.134(n) of the regulatory text.However, until the Department of Laborpublishes in the Federal Register thecontrol numbers assigned by the Officeof Management and Budget (OMB),affected parties are not required tocomply with the new or revisedinformation collection requirementscontained in the following paragraphs:§ 1910.134(c) written procedures forselecting respirators, medicalevaluations, fit testing, use ofrespirators, maintaining respirators,training, and periodically evaluating theeffectiveness of the program; (e)(3)–(6)medical questionnaire, examination,and information for the physician orother licensed health care professional(PLHCP); (f)(1) fit testing; (i)(4) taggingsorbent beds and filters; and (m)(1)–(2)and (4) recordkeeping. Publication ofthe control numbers notifies the publicthat the OMB has approved theseinformation collection requirementsunder the Paperwork Reduction Act of1995. Although affected parties will nothave to comply with the revisedstandard’s information collectionrequirements until these have beenapproved by OMB, they must complywith those requirements of 29 CFR1910.134 (OSHA’s existing respiratorprotection standard) that have alreadybeen approved by the OMB under thePaperwork Reduction Act. Approvedrequirements include the writtenprogram, emergency-use respiratorcertification records, and emergency-userespirator compartment marking.

Comments: Interested parties maysubmit comments on the informationcollection requirements for thisstandard until March 9, 1998.

ADDRESSES: In compliance with 28U.S.C. 2112(a), the Agency designatesthe Associate Solicitor for OccupationalSafety and Health, Office of theSolicitor, Room S–4004, U.S.Department of Labor, 200 ConstitutionAvenue, N.W., Washington, D.C. 20210,

as the recipient of petitions for reviewof the standard.

Comments on the informationcollection requirements of this final rule(see SUPPLEMENTARY INFORMATION) are tobe submitted to the Docket Office,Docket No. ICR 97–5, U.S. Departmentof Labor, Room N–2625, 200Constitution Avenue, N.W.,Washington, D.C. 20210, telephone(202) 219–7894. Written commentslimited to 10 pages or less in length mayalso be transmitted by facsimile to (202)219–5046.

Copies of the referenced informationcollection request are available forinspection and copying in the DocketOffice and will be mailed immediatelyto persons who request copies bytelephoning Adrian Corsey at (202) 219–7075. For electronic copies of theRespiratory Protection Final Standardand the Information Collection Request,contact OSHA’s WebPage on theInternet at http://www.osha.gov/.FOR FURTHER INFORMATION CONTACT:Bonnie Friedman, Director, OSHAOffice of Public Affairs, Room N–3647,U.S. Department of Labor, 200Constitution Avenue, N.W.,Washington, D.C. 20210; Telephone(202) 219–8148. For additional copies ofthis regulation contact: OSHA, Office ofPublications, U.S. Department of Labor,Room N–3101, 200 ConstitutionAvenue, N.W., Washington, D.C. 20210;Telephone (202) 219–4667.

SUPPLEMENTARY INFORMATION:

1. Collection of Information: Requestfor Comment

This final Respiratory Protectionstandard contains informationcollection requirements that are subjectto review by OMB under the PaperworkReduction Act of 1995 (PRA95), 44U.S.C. 3501 et seq. (see also 5 CFR1320). PRA95 defines collection ofinformation to mean, ‘‘the obtaining,causing to be obtained, soliciting, orrequiring the disclosure to third partiesor the public of facts or opinions by orfor an agency regardless of form orformat.’’ [44 U.S.C. § 3502(3)(A)]

The title, the need for and proposeduse of the information, a summary of thecollections of information, descriptionof the respondents, and frequency ofresponse required to implement therequired information collection aredescribed below with an estimate of theannual cost and reporting burden (asrequired by 5 CFR 1320.5 (a)(1)(iv) and§ 1320.8 (d)(2)). Included in the estimateis the time for reviewing instructions,gathering and maintaining the dataneeded, and completing and reviewingthe collection of information.

1153Federal Register / Vol. 63, No. 5 / Thursday, January 8, 1998 / Rules and Regulations

OSHA invites comments on whetherthe proposed collection of information:

• Ensures that the collection ofinformation is necessary for the properperformance of the functions of theagency, including whether theinformation will have practical utility;

• Estimates the projected burdenaccurately, including whether themethodology and assumptions used arevalid;

• Enhances the quality, utility, andclarity of the information to becollected; and

• Minimizes the burden of thecollection of information on those whoare to respond, including the use ofappropriate automated, electronic,mechanical, or other technological

collection techniques or other forms ofinformation technology, e.g., permittingelectronic submissions of responses.

Title: Respiratory Protection, 29 CFR1910.134.

Description: The final RespiratoryProtection standard is an occupationalhealth standard that will minimizeoccupational exposure to toxicsubstances. The standard’s informationcollection requirements are essentialcomponents that will protect employeesfrom occupational exposure to thesetoxins. The information will be used byemployers and employees to implementthe protection required by the standard.OSHA will use some of the informationto determine compliance with thestandard.

Respondents: The total number ofrespondents for the first year is1,300,000, and for the second year1,430,000 (1,300,000 (1st year) plus 10%(130,000)).

Average Time Per Response: 2.21hours (this is the result of dividing thetotal number of responses (19,767,461)by the total number of burden hours(8,926,558)).

Average Time Per Firm: 6.87 hours(this represents the average time a firmwould need to comply with all of theinformation collection provisions,including the written respiratoryprotection program. This is a result ofdividing the total number of burdenhours (8,926,558) by the total number offirms (1,300,000)).

SUMMARY OF THE COLLECTIONS OF INFORMATION

Information collectionrequirement

No. ofresponses

(Yr 1)

No. ofresponses

(Yr 2)

Frequency of re-sponse

Time perresponse

Total 1st yearburden

Estimatedcost

(1st year)

Respiratory Protection Program1910.134(c).

1,274,000 26,000 All Existing Firmsto Update Exist-ing Program.

2 Hours for SmallFirms; 4 Hoursfor Large Firms.

2,652,000 $60,916,440

127,400 2,600 Initially for NewEmployers.

8 Hours to De-velop.

Updates (Every 5Years).

30 Minutes forSmall Firms; 1Hour for LargeFirms.

Questionnaire Administration1910.134(e)(3).

5,000,000 575,000 All Employees WillReceive in theFirst Year.

50% of those Re-ceiving ExamsWill ReceiveFollow-up Ques-tionnaires.

15 Minutes forEmployees toComplete.

740,000 $13,593,800

Medical Examinations 1910.134(e)(4) 1,150,000 287,500 23% of the Exist-ing Employees.

2nd & RecurringYrs—25% of the23% would re-ceive Follow-upExams.

All Medical Examswill Take 1.5Hours to Com-plete which in-cludes traveltime.

1,021,200 $18,759,444

Information Provided to PLHCP1910.134(e)(5).

1,150,000 287,500 Dependent on theNumber ofExams.

15 Minutes forEach Employee.

170,200 $2,358,972

Fit Testing 1910.134(f)(1) .................... 4,335,000 4,335,000 346,800 Employ-ees to ReceiveQuantitative FitTests.

799,640 Employ-ees to ReceiveQualitative FitTests.

3,188,560 Employ-ees to ReceiveIn-House FitTests.

4,335,000 TotalEmployees.

30 Minutes forEmployees tobe Fitted (Quan-titative andQualitative FitTesting).

30 Additional Min-utes for Employ-ers to Conduct(Only for In-House Fit Test-ing).

3,780,140 $76,813,315

Emergency-Use Respirator Marking1910.134(h)(2)(ii)(B).

0 260,000 Only New Employ-ers E.

xisting EmployersHave AlreadyComplied (OldRequirement).

5 Minutes perEmergency-UseRespirator.

0 $0


SUMMARY OF THE COLLECTIONS OF INFORMATION—Continued


No. ofresponses

(Yr 1)

No. ofresponses

(Yr 2)

Frequency of re-sponse

Time perresponse

Total 1st yearburden

Estimatedcost

(1st year)

Emergency-Use Respirator Certifi-cation 1910.134(h)(3)(iv)(A)&(B).

671,880 67,200 Currently, 27,995EmployersUsing Emer-gency-Use Res-pirators (1stYear).

2nd Year = 1stYear Employersplus 10%.

Assuming 2 PerEmployer: 10Minutes (TotalTime PerMonth).

114,220 $2,098,221

Certificate of Analysis of Cylinders1910.134(i)(4)(i)(B).

0 0 All Existing andNew Employers.

Provided by Sup-plier, thereforeno burden in-curred.

0 $0

Sorbent Beds and Filters1910.134(i)(4)(iii)(B).

74,181 74,181 Currently, 24,727Compressors inUse.

3 Changes PerYear, assuming5 minutes perchange.

5,934 $109,008

Medical Records 1910.134(m)(1) ........ 1,150,000 287,500 Dependent on theNumber ofExams.

5 Minutes Per Em-ployee Exam-ined.

54,464 $754,871

Fit Testing Records 1910.134(m)(2) .... 4,335,000 4,335,000 Dependent on theNumber of FitTests.

5 Minutes Per FitTest.

348,400 $4,828,824

Employee Access 1910.134(m)(4) ...... 500,000 500,000 10% of the TotalNumber of Em-ployees.

5 Minutes per Re-quest.

40,000 $554,400

Totals ........................................ 19,767,461 11,037,481 ......................... ......................... 8,926,558 $180,787,295

MARGINAL DIFFERENCES IN BURDEN HOURS AND COSTS (I.E., BETWEEN THE EXISTING AND REVISED STANDARDS)


Current OMBinventory ex-

isting1910.134

Adjustment (to1st year only)

1st yr. burdenrevised

1910.134Estimated cost

2nd & recur-ring yr. burden

revised1910.134

Estimated cost

Respiratory Protection Program ............... 395,489 2,256,511 2,652,000 $60,916,440 1,570,400 $36,072,088Questionnaire Administration .................... – 740,000 740,000 $13,593,800 85,100 $1,563,287Medical Examinations ............................... – 1,021,200 1,021,200 $18,759,444 255,300 $4,689,861Information Provided to PLHCP ............... – 170,200 170,200 $2,358,972 42,550 $589,743Fit Testing ................................................. – 3,780,140 3,780,140 $76,813,315 3,780,140 $76,813,315Emergency-Use Respirator Marking ......... 433 –433 0 $0 448 $8,230Emergency-Use Respirator Certification .. 785,842 –671,622 114,220 $2,098,221 11,424 $209,859Certificate of Analysis of Cylinders ........... – 0 0 $0 0 $0Sorbent Beds and Filters .......................... – 5,934 5,934 $109,008 5,934 $109,008Medical Records ....................................... – 54,464 54,464 $754,871 13,616 $188,718Fit Testing Records ................................... – 348,400 348,400 $4,828,824 348,400 $4,828,824Employee Access ..................................... – 40,000 40,000 $554,400 40,000 $554,400Hour Kept in Inventory for Revised

1910.134 ................................................ 1 –1 0 $0 0 $0

Totals .............................................. 1,181,765 7,744,793 8,926,558 $180,787,295 6,153,312 $125,627,333

Under the column for ‘‘Current OMB Inventory,’’ dashes denote burdens that were not taken for the Existing Respiratory Protection Standard,but are counted in the Revised Respiratory Protection Standard. Both Medical Examinations and Fit Testing are required by the existing stand-ard; however, because these requirements are not accompanied by a recordkeeping requirement, no burden was taken. In the revised standard,recordkeeping is required for these provisions, and thus burden is counted for these provisions.

Interested parties are requested tosend comments regarding thisinformation collection to the OSHADocket Office, Docket No. ICR 97–5 ,U.S. Department of Labor, Room N–2625, 200 Constitution Avenue, N.W.,Washington, D.C. 20210. Writtencomments limited to 10 pages or fewermay also be transmitted by facsimile to(202) 219–5046.

Comments submitted in response tothis notice will be summarized andincluded in the request for Office ofManagement and Budget approval of thefinal information collection request;they will also become a matter of publicrecord.

Copies of the referenced informationcollection request are available forinspection and copying in the OSHA

Docket Office and will be mailed topersons who request copies bytelephoning Adrian Corsey at (202) 219–7075. Electronic copies of theRespiratory Protection Final informationcollection request are available on theOSHA WebPage on the internet at http://www.osha.gov/ under Standards.


2. Federalism

This final standard has been reviewedin accordance with Executive Order12612 (52 FR 41685, October 30, 1987),regarding Federalism. This Orderrequires that agencies, to the extentpossible, refrain from limiting statepolicy options, consult with states priorto taking any actions which wouldrestrict state policy options, and takesuch actions only when there is clearconstitutional authority and thepresence of a problem of national scope.The Order provides for preemption ofstate law only if there is a clearCongressional intent for the Agency todo so. Any such preemption is to belimited to the extent possible.

Section 18 of the Occupational Safetyand Health Act (OSH Act) expressesCongress’ clear intent to preempt statelaws relating to issues on which FederalOSHA has promulgated occupationalsafety and health standards. Under theOSH Act, a state can avoid preemptiononly if it submits, and obtains Federalapproval of, a plan for the developmentof such standards and theirenforcement. Occupational safety andhealth standards developed by suchPlan-States must, among other things, beat least as effective in providing safe andhealthful employment and places ofemployment as the Federal standards.Where such standards are applicable toproducts distributed or used ininterstate commerce, they may notunduly burden commerce and must bejustified by compelling local conditions(see OSH Act, Section 18(c)).

The final Federal standard onrespiratory protection addresses hazardswhich are not unique to any one stateor region of the country. Nonetheless,states with occupational safety andhealth plans approved under Section 18of the OSH Act will be able to developtheir own state standards to deal withany special problems which might beencountered in a particular state.Moreover, because this standard iswritten in general, performance-orientedterms, there is considerable flexibilityfor state plans to require, and foraffected employers to use, methods ofcompliance which are appropriate to theworking conditions covered by thestandard.

In brief, this final standard addressesa clear national problem related tooccupational safety and health ingeneral industry, construction, andmaritime employment. Those stateswhich have elected to participate underSection 18 of the OSH Act are notpreempted by this standard, and will beable to address any special conditionswithin the framework of the Federal Act

while ensuring that the state standardsare at least as effective as that standard.

3. State PlansThe 25 states and territories with their

own OSHA-approved occupationalsafety and health plans must adopt acomparable standard within six monthsof the publication date of a finalstandard. These 25 states are: Alaska,Arizona, California, Connecticut, NewYork (for state and local governmentemployees only), Hawaii, Indiana, Iowa,Kentucky, Maryland, Michigan,Minnesota, Nevada, New Mexico, NorthCarolina, Oregon, Puerto Rico, SouthCarolina, Tennessee, Utah, Vermont,Virginia, Virgin Islands, Washington,and Wyoming. Until such time as a statestandard is promulgated, Federal OSHAwill provide interim enforcementassistance, as appropriate, in thesestates.

4. Unfunded MandatesThe final respiratory protection rule

has been reviewed in accordance withthe Unfunded Mandates Reform Act of1995 (UMRA) (2 U.S.C. 1501 et seq.) andExecutive Order 12875. As discussedbelow in the Summary of the FinalEconomic Analysis (FEA) (Section VI ofthis document), OSHA estimates thatcompliance with the revised respiratoryprotection standard will require theexpenditure of more than $100 millioneach year by employers in the privatesector. Therefore, the final ruleestablishes a Federal private sectormandate and is a significant regulatoryaction, within the meaning of section202 of UMRA (2 U.S.C. 1532). OSHAhas included this statement to addressthe anticipated effects of the finalrespiratory protection rule pursuant tosection 202.

OSHA standards do not apply to stateand local governments, except in statesthat have voluntarily elected to adopt anOSHA State Plan. Consequently, therespiratory protection standard does notmeet the definition of a ‘‘Federalintergovernmental mandate’’ (Section421(5) of UMRA (2 U.S.C. 658(5)). Thus,the final respiratory protection standarddoes not impose unfunded mandates onstate or local governments.

The anticipated benefits and costs ofthis final standard, and other issuesraised in section 202 of the UMRA, areaddressed in the Summary of the FEA(Section VI of this preamble), below,and in the FEA (Ex. 196). In addition,pursuant to section 205 of the UMRA (2U.S.C. 1535), having considered areasonable number of alternatives asoutlined in the preambles to theproposal and the final rule and in theFEA (Ex. 196), the Agency has

concluded that the final rule is the mostcost-effective alternative forimplementation of OSHA’s statutoryobjective of reducing significant risk tothe extent feasible. This is discussed inthe FEA (Ex. 196) and in the Summaryand Explanation (Section VII of thispreamble) for the various provisions ofthe final standard.

5. Executive Order 13045—Protection ofChildren From Environmental Healthand Safety Risks

Executive Order 13045, signed by thePresident on April 21, 1997, requiresthat for certain Federal agency‘‘regulatory actions submitted to OMB’sOffice of Information and RegulatoryAffairs (OIRA) for review pursuant toExecutive Order 12866, the issuingagency shall provide to OIRA thefollowing information developed as partof the Agency’s decisionmaking process,unless prohibited by law:

(a) An evaluation of theenvironmental health or safety effects ofthe planned regulation on children; and

(b) An explanation of why theplanned regulation is preferable to otherpotentially effective and reasonablyfeasible alternatives considered by theagency.’’

‘‘Covered Regulatory Actions’’ underthis Order are rules that may:

(a) Be ‘‘economically significant’’under Executive Order 12866 (arulemaking that has an annual effect onthe economy of $100 million or more orwould adversely affect in a material waythe economy, a sector of the economy,productivity, competition, jobs, theenvironment, public health or safety, orState, local, or tribal governments orcommunities); and

(b) Concern an environmental healthrisk or safety risk that an agency hasreason to believe maydisproportionately affect children.

‘‘Environmental health risks andsafety risks’ mean risks to health or tosafety that are attributable to products orsubstances that the child is likely tocome in contact with or ingest (such asthe air we breathe, the food we eat, thewater we drink or use for recreation, thesoil we live on, and the products we useor are exposed to).

The final standard on respiratoryprotection does not concern‘‘Environmental health risks and safetyrisks’’ to children as defined under theExecutive order. The respirator standardis only concerned with means oflimiting employee exposures to toxicsubstances. The Agency believes,therefore, that the requirement notedabove to provide OIRA with certaininformation does not apply since therespiratory protection standard is not a


‘‘covered regulatory action’’ underExecutive Order 13045.

Section 6(b) (8) of the OSH Actrequires OSHA to explain ‘‘why a rulepromulgated by the Secretary differssubstantially from an existing nationalconsensus standard,’’ by publishing ‘‘astatement of the reasons why the rule asadopted will better effectuate thepurposes of the Act than the nationalconsensus standard.’’ In compliancewith the requirement, the Agency hasreviewed the standards proposedthrough this rulemaking with referenceto the ANSI Z88.2–1992 standard forRespiratory Protection. OSHA hasdiscussed the relationship betweenindividual regulatory provisions and thecorresponding consensus standards inthe Summary and Explanation of thefinal rule.

6. Reasons Why the Revised Rule WillBetter Effectuate the Purposes of the ActThan the Existing Consensus Standard

This process was facilitated by thefact that the previous OSHA standardson respiratory protection were start-upstandards adopted directly from theANSI Z88.2–1969 standard, ‘‘Practicesfor Respiratory Protection’’ undersection 6(a) of the OSH Act, 29 U.S.C.655(a). Therefore, even with subsequentrevisions to the ANSI standards and theAgency’s consideration of a widelyvaried and substantial body ofinformation in the rulemaking record,the requirements of the OSHA final rulewould tend to resemble thecorresponding provisions of the currentANSI standards. In a number ofinstances, OSHA has utilized languageidentical to that in the current ANSIstandard. These instances are noted inthe Summary and Explanation. Wherethe Agency has determined that thepertinent ANSI language is notappropriate for this OSHA standard, theSummary and Explanation provides thebasis for that decision.

I. GeneralThe preamble accompanying this final

standard discusses events leading to thefinal rule, the types of respiratoryhazards experienced by employees, thedegree and significance of the riskpresented by failure to comply with thisrevised standard, the Final EconomicAnalysis, and the rationale behind thespecific provisions set forth in the finalstandard. The discussion follows thisoutline:I. GeneralII. Pertinent Legal AuthorityIII. Events Leading to the Final Standard

A. Regulatory HistoryB. Justification for Revising the Previous

Standard

1. Purpose of Revision2. Respirator Use and HazardsC. Responses to Advisory CommitteeD. Assigned Protection FactorsE. Small Business Considerations

IV. Certification/Approval ProceduresV. Significance of RiskVI. Summary of the Final Economic Analysis

And Environmental Impact AssessmentVII. Summary And Explanation of the Final

StandardA. Permissible PracticeB. DefinitionsC. Respiratory Protection ProgramD. Selection of RespiratorsE. Medical EvaluationF. Fit Testing ProceduresG. Use of RespiratorsH. Maintenance and Care of RespiratorsI. Breathing Air Quality and UseJ. Identification of Filters, Cartridges, and

CanistersK. TrainingL. Respiratory Protection Program

EvaluationM. Recordkeeping and Access to RecordsN. DatesO. AppendicesP. Revisions to Specific Standards

VIII. Authority And SignatureIX. Amended Standards

II. Pertinent Legal AuthorityThe purpose of the Occupational

Safety and Health Act, 29 U.S.C. 651 etseq. (‘‘the Act’’) is to ‘‘assure so far aspossible every working man and womanin the nation safe and healthful workingconditions and to preserve our humanresources.’’ 29 U.S.C. 651(b). To achievethis goal, Congress authorized theSecretary of Labor to promulgate andenforce occupational safety and healthstandards. U.S.C. 655(a) (authorizingsummary adoption of existingconsensus and Federal standards withintwo years of Act’s enactment), 655(b)(authorizing promulgation of standardspursuant to notice and comment),654(b) (requiring employers to complywith OSHA standards).

A safety or health standard is astandard ‘‘which requires conditions, orthe adoption or use of one or morepractices, means, methods, operations,or processes, reasonably necessary orappropriate to provide safe or healthfulemployment or places of employment.’’29 U.S.C. 652(8).

A standard is reasonably necessary orappropriate within the meaning ofsection 652(8) if it substantially reducesor eliminates significant risk or preventsit from developing, and is economicallyfeasible, technologically feasible, costeffective, consistent with prior Agencyaction or supported by a reasonedjustification for departing from priorAgency actions, supported bysubstantial evidence, and is better ableto effectuate the Act’s purposes than anynational consensus standard it

supersedes. See 58 FR 16612–16616(March 30, 1993).

A standard is technologically feasibleif the protective measures it requiresalready exist, can be brought intoexistence with available technology, orcan be created with technology that canreasonably be expected to be developed.American Textile Mfrs. Institute v.OSHA, 452 U.S. 490, 513 (1981)(‘‘ATMI’’), American Iron and SteelInstitute v. OSHA, 939 F.2d 975, 980(D.C. Cir. 1991)(‘‘AISI’’).

A standard is economically feasible ifindustry can absorb or pass on the costof compliance without threatening itslong term profitability or competitivestructure. See ATMI, 452 U.S. at 530 n.55; AISI, 939 F. 2d at 980.

A standard is cost effective if theprotective measures it requires are theleast costly of the available alternativesthat achieve the same level ofprotection. ATMI, 453 U.S. at 514 n. 32;International Union, UAW v. OSHA, 37F.3d 665, 668 (D.C. Cir. 1994)(‘‘LOTOIII’’).

All standards must be highlyprotective. See 58 FR 16614–16615;LOTO III, 37 F.3d at 668. However,standards regulating exposure to toxicsubstances or hazardous physical agentsmust also meet the ‘‘feasibilitymandate’’ of Section 6(b)(5) of the Act,29 U.S.C. 655(b)(5). Section 6(b)(5)requires OSHA to select ‘‘the mostprotective standard consistent withfeasibility’’ that is needed to reducesignificant risk when regulating thesehazards. ATMI, 452 U.S. at 509.

Section 6(b)(5) also directs OSHA tobase health standards on ‘‘the bestavailable evidence,’’ including research,demonstrations, and experiments, 29U.S.C. 655(b)(5). OSHA shall consider‘‘in addition to the attainment of thehighest degree of health and safetyprotection * * * the latest scientificdata * * * feasibility and experiencegained under this and other health andsafety laws.’’ Id.

Section 6(b)(7) of the Act authorizesOSHA to include among a standard’srequirements labeling, monitoring,medical testing and other informationgathering and transmittal provisions. 29U.S.C. 655(b)(7).

Finally, whenever practical, standardsshall ‘‘be expressed in terms of objectivecriteria and of the performancedesired.’’ Id.

Respiratory protection is a backupmethod which is used to protectemployees from toxic materials in theworkplace in those situations wherefeasible engineering controls and workpractices are not available, have not yetbeen implemented, are not inthemselves sufficient to protect


employee health, or in emergencies. Therevisions to the respirator standardmade in this rulemaking are intended toensure that, when employers requireemployees to wear respirators to beprotected from significant risk,protective respirators will be selectedand those respirators will be usedeffectively to meet their designcapabilities. Otherwise respirators willnot reduce significant risk. Thestandard’s provisions are designed to befeasible and cost effective, and areexpressed in terms of objective criteriaand the performance desired.

Further authority is provided bysection 8(c)of the Act, which authorizesOSHA to require employers to maintaincertain records. Section 8(g)(2)authorizes OSHA ‘‘to prescribe suchrules and regulations as (it) may deemnecessary to carry out itsresponsibilities under the Act.’’

III. Events Leading to the FinalStandard

A. Regulatory History

Congress created the OccupationalSafety and Health Administration(OSHA) in 1970, and gave it theresponsibility for promulgatingstandards to protect the health andsafety of American workers. As directedby Congress in the Occupational Safetyand Health Act of 1970 (OSH Act; 29U.S.C. 651 et seq.), OSHA adoptedexisting Federal standards and nationalconsensus standards developed byvarious organizations such as theAmerican Conference of GovernmentalIndustrial Hygienists (ACGIH), theNational Fire Protection Association(NFPA), and the American NationalStandards Institute (ANSI). The ANSIstandard Z88.2–1969, ‘‘Practices forRespiratory Protection,’’ is the basis ofthe first six sections of OSHA’s previousstandard, 29 CFR 1910.134,‘‘Respiratory Protection.’’ The seventhsection was a direct, completeincorporation of ANSI Standard K13.1–1969, ‘‘Identification of Gas MaskCanisters.’’ OSHA’s previousconstruction industry standard forrespiratory protection, 29 CFR 1926.103,was promulgated in April 1971. OnFebruary 9, 1979, 29 CFR 1910.134 wasformally recognized as also beingapplicable to the construction industry(44 FR 8577). Until the adoption ofthese standards by OSHA, mostguidance on respiratory protectivedevice use in hazardous environmentswas advisory rather than mandatory.

OSHA’s maritime standards wereoriginally promulgated in the 1960s byagencies that preceded OSHA. Theoriginal OSHA code designations of

these standards and their promulgationdates are: Shipyards—29 CFR 1915.82,February 20, 1960 (25 FR 1543); MarineTerminals—29 CFR 1917.82, March 27,1964 (29 FR 4052); and Longshoring—29 CFR 1918.102, February 20, 1960 (25FR 1565). Section 1910.134 wasincorporated by reference into OSHA’sMarine Terminals standard (part 1917)on July 5, 1983 (48 FR 30909). OSHAhas recently updated and strengthenedits Longshoring and Marine Terminalstandards, and both standardsincorporate 29 CFR 1910.134 byreference.

OSHA did not propose to expandcoverage of 29 CFR 1910.134 toagricultural workplaces covered by 29CFR part 1928, and this finalRespiratory Protection standard, like theproposal, does not apply to agriculturaloperations. The prior standard likewisedid not apply to agricultural operations.(See 29 CFR 1928.21.) OSHA receivedno public comment requesting a changein coverage. Accordingly, the issue ofrespirator use during agriculturaloperations was not a part of thisrulemaking. OSHA notes, however, thatrespirator use during pesticideoperations and handling is covered byEPA’s Worker Protection Standard, 40U.S.C. part 170, adopted under theauthority of the Federal Insecticide,Fungicide, and Rodenticide Act, asamended (7 U.S.C. 136–136y).

Under OSHA’s previous standard,employers needed to follow theguidance of the Z88.2–1969 ANSIstandard to ensure proper selection ofrespirators (see discussion 59 FR58887). OSHA published an AdvanceNotice of Proposed Rulemaking (ANPR)to revise the respirator standard on May14, 1982 (47 FR 20803). Part of theimpetus for this notice was OSHA’sinclusion of new respiratorrequirements in comprehensivesubstance-specific standardspromulgated under section 6(b) of theAct, e.g., fit tests; use of powered air-purifying respirators (PAPRs) uponrequest; change of the filter elements ofa respirator whenever an increase inbreathing resistance is detected;employee permission to wash faces andrespirator facepieces; and referral to aphysician trained in pulmonarymedicine for an employee who exhibitsdifficulty breathing, either at fit testingor during routine respirator use (see, e.g,29 CFR 1910.1025 (lead standard)). Therespirator provisions in these substance-specific standards took account ofadvances in respirator technology andchanges in related guidance documents,particularly the recognition thatstandardized fit testing protocols greatlyincrease the effectiveness of respirators.

OSHA’s 1982 ANPR soughtinformation on the effectiveness of thecurrent respiratory protectionprovisions, the need for revision ofthose provisions, and the substance ofthe revisions. Responses were receivedfrom 81 interested parties. Thecommenters generally supportedrevising OSHA’s respiratory protectionprovisions and provided suggestions forapproaches the Agency might take (Ex.15).

On September 17, 1985, OSHAannounced the availability of apreliminary draft of the proposedRespiratory Protection standard. Thepreproposal draft standard reflected thepublic comments received on the May1982 ANPR, and OSHA’s own analysisof changes needed in the standard totake into account the current state-of-the-art for respiratory protection.Responses were received from 56interested parties (Ex. 36), and theircomments were reviewed in preparingthe proposal.

On November 15, 1994, OSHApublished the proposed rule to revise 29CFR 1910.134, and announced itsintention to convene an informal publichearing on the proposal (59 FR 58884).The informal public hearing wasconvened on June 6, 1995, pursuant tonotice and in accordance with Section6(b) of the OSH Act, 29 U.S.C. 655(b)(3).Post-hearing submissions of data fromparties at the hearing were receivedthrough September 20, 1995.

On November 7, 1995, OSHAreopened the record (60 FR 56127) andrequested additional comment on astudy performed for OSHA by Dr. MarkNicas titled ‘‘The Analysis of WorkplaceProtection Factor Data and Derivation ofAssigned Protection Factors.’’ Thatstudy, which was placed in therulemaking docket on September 20,1995, addressed the use of statisticalmodeling for determining respiratorAPFs. Comments on the Nicas studywere received through the end ofJanuary 1996. The Nicas report, andcomments received in response to theNovember 1995 notice, have convincedOSHA to deliberate further on thecomplex issues surrounding theestablishment of APFs.

The entire record including 200exhibits, more than 3,000 individualitems, and approximately 2,300transcript pages, was certified by thepresiding administrative law judge onJune 30, 1997, in accordance with 29CFR 1911.17. Copies of materialscontained in the record may be obtainedfrom the OSHA Docket Office, Room N–2439, U.S. Department of Labor, 200Constitution Avenue, N.W.,


Washington, D.C. 20210; (202) 219–7894.

The final revisions to 29 CFR1910.134 are based on consideration ofthe entire record of this proceeding,including materials discussed or reliedupon in the proposal, the record of theinformal hearing, and all writtencomments and exhibits received.

B. Justification for Revising the PreviousStandard

1. Purpose of the Revision

The intent of this revision is toenhance the protection of worker health,promote more effective use ofrespirators, provide greater complianceflexibility, and clarify the policies andprocedures employers must followwhen implementing a respiratoryprotection program. Evidence in therecord, including case reports andstudies of respirator use among workers,indicates that selecting or usingrespirators improperly can result inemployee illness and even death. (Seediscussion below.) The revised standardis therefore expected to reduce thenumber of occupational illnesses anddeaths among workers who wearrespirators. OSHA is also consolidatingmany of its respirator-related provisionsin other substance-specific healthstandards into one standard to makethese provisions easier for employers toadminister. Through consolidation,repetitive and duplicative respiratorrequirements have been deleted frommany existing OSHA health standards,and future health standards willreference the revised final rule for manyrespirator requirements.

Advances in technology also made theprevious standard out-of-date in manyareas. Nearly all rulemakingparticipants, including representativesof private industry, other Federalagencies, respirator manufacturers, andunions, agreed that revision is necessaryto address these advances (e.g., NIOSH,Ex. 28; Eastman Chemical Co., Ex. 54–245; 3M, Ex. 54–218A; AFL–CIO, Ex.54–315; Building and ConstructionTrades Department/AFL–CIO, Ex. 29;American Petroleum Institute, Ex. 37;ISEA, Ex. 54–363). (See also 59 FR58889.) Other agencies and committeeshave already updated their guidance onrespirator use. For example, the ANSIstandard has been revised twice (Exs.10, 50), and NIOSH has revised itscertification standard (42 CFR part 84;60 FR 30336; 6/8/95), as well asdeveloped a Respiratory Decision Logic(1987) to provide guidance to employerson the selection of respirators.

OSHA’s experience in enforcing theprevious standard also indicated that

some of that standard’s requirementswere not understood clearly by theregulated community, and so were notadequately effective in protectingworkers. The clarifications in this newstandard will contribute to enhancedcompliance by reducingmisinterpretations and inconsistencies.A review of OSHA enforcement data for1994 and 1995 revealed that failure tocomply with the previous standard wasa critical factor in at least 47 fatalitiesand 126 catastrophic injuries. The mostfrequently cited deficiencies includedfailure to provide respirators at all or tohave standard operating proceduresgoverning respirator use, and failure totrain or fit test respirator usersadequately [Source: OSHA’s FederalInspection Compliance Data (IMIS; 10/92 to 12/95)].

In addition, considerable research hasbeen performed to determine the extentto which respirators used in workplacesactually reduce the quantity ofcontaminant breathed by the respiratoruser. Researchers have compared the in-mask concentrations of contaminants tothe concentration levels outside themasks. This work was begun by NIOSHduring the mid-seventies to assessrespirator effectiveness in coal minesand abrasive blasting operations (Ex.64–5) and spray paint operations (Ex.64–68). The studies assessed theeffectiveness of respirators undervarious conditions, and measuredemployee exposure in situations whenrespirators were not worn. Theeffectiveness ratings obtained in thesestudies are usually termed ‘‘EffectiveProtection Factors’’ (EPF).

More recent studies by NIOSH andprivate researchers have monitoredrespirator use even more closely toisolate variables that may affect thelevels of respirator performance. Manyof these studies concerned theperformance of powered air-purifyingrespirators (PAPRs), which were notachieving in workplaces the levels ofperformance that had been predictedbased on laboratory tests (see, e.g., Exs.64–46, 64–42, and 64–47).

A third group of studies, ‘‘workplaceprotection factor studies,’’ conductedmostly by manufacturers and otherprivate interests, was designed todetermine the optimum performance ofrespirators by eliminating the impact ofprogram defects under very tightlysupervised workplace conditions. Theresults of these studies may overstatethe degree of respirator effectivenessmost employers can expect underconditions of workplace use becausestudy conditions are rarely replicated inthe field; nevertheless, these studiesshow the potential for respirators to

reduce employee exposure to workplacecontaminants (see, e.g., Exs. 64–25, 64–42, 64–47, 64–513).

This revised standard is intended totake account of up-to-date knowledgeand technology and to make therequirements in the standard easier tounderstand. The standard now reflectscurrent technology and research, as wellas the findings and guidance of otherexpert bodies. OSHA has also includeda new definitions section to enhanceclarity. The revised standard includesdetailed protocols for performing fittests and lists the topics in whichrespirator users must be trained. It alsocontains provisions addressing skin andeye irritation, both of which must beconsidered in respirator selection.Wherever possible, OSHA has usedperformance-oriented language to allowfor flexibility in accommodating futurechanges in respirator technology and toaddress the needs of small businessesand unusual operations. Through theseimprovements, OSHA expects to reducethe number of respirator-relatedillnesses, fatalities, and catastrophicinjuries occurring among respiratorwearers in U.S. workplaces.

2. Respirator Use and HazardsThe purpose of a respirator is to

prevent the inhalation of harmfulairborne substances or oxygen-deficientair. Basically, a respirator is anenclosure that covers the nose andmouth or the entire face or head.Respirators are of two general ‘‘fit’’types: (1) Tight-fitting (quarter masks,which cover the mouth and nose; halfmasks, which fit over the nose andunder the chin; and full facepiece,which cover the face from the hairlineto below the chin); and (2) loose-fitting(hoods, helmets, blouses, or full suitswhich cover the head completely).There are also two major classes ofrespirators: air-purifying respirators(which remove contaminants from theair), and atmosphere-supplyingrespirators (which provide cleanbreathing air from an uncontaminatedsource). In general, atmosphere-supplying respirators are used for morehazardous exposures.

Effective respirator use can protectemployees from exposure to a widevariety of toxic chemicals. In 1994,approximately 215 deaths, or fivepercent of all workplace fatalities,occurred as a result of exposure toharmful substances and environments[CFOI, BLS, 6/11/96; CFOI/FAX]. Thereare a number of workplace situationsthat involve toxic substances and forwhich engineering controls may beinadequate to control exposures, andrespirators are used in these situations


as a back-up method of protection.Substances that have been associatedwith death or serious incidents includecarbon monoxide, trichloroethylene,carbon dioxide, chromic acid, coal tar,several toxic metal fumes and dusts,sulphur dioxide, wood dust, andwelding fumes; these substances causeadverse health effects ranging fromtransient, reversible effects such asirritation or narcosis, through disablingdiseases such as silicosis and asbestosis,to death caused either by acute exposureor by a cancer resulting from chronicexposures (Rom, W., Environmental andOccupational Medicine, 2nd ed., Little,Brown & Co., Boston; 1992, p. 598.)Respirators are available that canprovide protection against inhalation ofthese toxic substances.

Airborne contaminants may also beradioactive (‘‘Radiologic Health inOccupational Medicine Practice,’’George L. Voelz, pg. 500 inOccupational Medicine, Carl Zenz, ed.,Year Book Medical Publishers, Inc.,Chicago, 1975; Jacob Shapiro, RadiationProtection, 3rd ed., Harvard UniversityPress, Cambridge, MA, 1990, pg. 273).(See also 29 CFR 1910.1096.) Exposureto ionizing radiation can cause acuteeffects such as nausea and vomiting,malaise and fatigue, increasedtemperature, and blood changes. Moresevere delayed effects include leukemia,bone and lung cancer, sterility,chromosomal and teratogenic damage,shortened life span, cataracts, andradiodermatitis, a dry, hairless, red,atrophic skin condition which caninclude skin cracking anddepigmentation (George L. Voelz, M.D.,‘‘Radiologic Health in OccupationalMedicine Practice’’, in Zenz,Occupational Medicine, pp. 513–519;Herman Cember, Introduction to HealthPhysics, 2nd edition, Pergamon Press,New York, 1983, pg. 181–194).Respirators to provide protection againstthe inhalation of radioactive particlesare commonly used by workers exposedto these hazards.

‘‘Bioaerosols’’ are airbornecontaminants that are alive or werereleased from a living organism (OSHADocket No. H–122; ACGIH Guidelines;Ex. 3–61C, page 1; 1994). Pulmonaryeffects associated with exposure tocertain bioaerosols include rhinitis,asthma, allergies, hypersensitivitydiseases, humidifier fever, andepidemics of infections including colds,viruses, tuberculosis, and LegionnairesDisease. Cardiovascular effectsmanifested as chest pain, and nervoussystem effects manifested as headache,blurred vision, and impaired judgment,have occurred in susceptible peoplefollowing exposure to bioaerosols. Viral

infections caused by the inhalation ofbioaerosols can result in health effectsthat range in intensity from undetectedor mild to more severe and even death.Bacterial infections resulting frominhalation of bacteria and their productscause a range of diseases, includingtuberculosis, Legionnaires Disease, andhypersensitivity pneumonitis. Amongworkers in sewage treatment plants,health-related problems can beassociated with occupational exposuresto protozoa [Burge, H., 1990,‘‘Bioaerosols: Prevalence and healtheffects in the indoor environment,’’ J.Allergy and Clinical Immunology; 86(5); see also Exs. 3–61B and 3–61C inDocket No. H–122.] Allergic asthma andallergic rhinitis can be induced bychronic exposure to low levels ofantigens. Hypersensitivity pneumonitiscan occur when a worker inhalesconcentrated aerosols of particlesreleased by bacteria, fungi, and protozoa(Exs. 3–61B and 3–61C in Docket No.H–122). In 1994, the Centers for DiseaseControl reported 41 deaths of workersfor which there was evidence of work-related hypersensitivity pneumonitis(Work-Related Lung DiseaseSurveillance Report, 1994; USDHHS,CDC, DHHS (NIOSH) Number 94–120).Respirators to protect against theinhalation of biological agents arewidely used in healthcare and otherworkplace settings where exposure tosuch agents presents a hazard toworkers.

Respirators can also provideprotection from oxygen-deficientatmospheres. Human beings mustbreathe oxygen in order to survive, andbegin to suffer adverse health effectswhen the oxygen level of their breathingair drops below the normal atmosphericlevel. Below 19.5 percent oxygen byvolume, air is considered oxygen-deficient. At concentrations of 16 to19.5 percent, workers engaged in anyform of exertion can rapidly becomesymptomatic as their tissues fail toobtain the oxygen necessary to functionproperly (Rom, W., Env. Occup. Med.,2nd ed; Little, Brown; Boston, 1992).Increased breathing rates, acceleratedheartbeat, and impaired thinking orcoordination occur more quickly in anoxygen-deficient environment. Even amomentary loss of coordination may bedevastating to a worker if it occurswhile the worker is performing apotentially dangerous activity, such asclimbing a ladder. Concentrations of 12to 16 percent oxygen cause tachypnea(increased breathing rates), tachycardia(accelerated heartbeat), and impairedattention, thinking, and coordination

(e.g., Ex. 25–4), even in people who areresting.

At oxygen levels of 10 to 14 percent,faulty judgment, intermittentrespiration, and exhaustion can beexpected even with minimal exertion(Exs. 25–4 and 150). Breathing aircontaining 6 to 10 percent oxygenresults in nausea, vomiting, lethargicmovements, and perhapsunconsciousness. Breathing aircontaining less than 6 percent oxygenproduces convulsions, then apnea(cessation of breathing), followed bycardiac standstill. These symptomsoccur immediately. Even if a workersurvives the hypoxic insult, organs mayshow evidence of hypoxic damage,which may be irreversible (Exs. 25–4and 150; also reported in: Rom, W.,Environmental and OccupationalMedicine, 2nd ed; Little, Brown; Boston,1992).

A number of workplace conditionscan lead to oxygen deficiency. Simpleasphyxiants, or gases that arephysiologically inert, can causeasphyxiation when present in highenough concentrations to lower theoxygen content in the air. Other toxic orchemical asphyxiants poisonhemoglobin, cytochromes, or otherenzyme systems (Rom, W.,Environmental and OccupationalMedicine, 2nd ed., Little, Brown, andCo., Boston, 1992). A number ofasphyxiants are gases that can evolvefrom explosions, combustion, chemicalreactions, or heating. A high-temperature electrical fire or arcwelding accident causing a completeflashover in an enclosed area cantemporarily eliminate oxygen from thatarea. Asphyxiation and the severe lungdamage it can cause are major concernsfor firefighters; of 30 firefighter deathsinvestigated by OSHA recently, fiveresulted from either asphyxiation,smoke inhalation, or flashovers (IMIS; 8State plan states; 10/91–3/97). (See alsomortality study of causes of deathamong firefighters, Guidotti, 37 JOEM1348, 1995.)

In 1994, 110 employees died fromoxygen deficiency [National Census ofFatal Occupational Injuries (CFOI); BLS;CFOI/FAX; 6/11/96)], i.e., about twopercent of the total number ofemployees who died of occupationalinjuries. OSHA believes that many ofthese deaths could have been preventedif the victims’ employers had realizedthat respirators were needed (BLS;CFOI/FAX, 6/96).

In some cases, respirator use itself cancause illness and injury to employees.There are a number of physiologicalburdens that are associated with the useof certain types of respirators. The


weight of the respirator, breathingresistances during both normaloperation and if the air-purifyingelement is overloaded, and rebreathingexhaled air from respirator ‘‘deadspace’’ can all increase the physiologicburden of respirator use (Exs. 113, 22–1, 64–427). Job and workplaceconditions, such as the length of time arespirator must be worn, the level ofphysical exertion required of arespirator user, and environmentalconditions, can also affect thephysiological burden (Exs. 113, 64–363).In addition, workers who wear glassesor hearing aids may have problemsachieving appropriate fit with somerespirator facepieces.

Evidence of Adverse Health EffectsFrom Respiratory Hazards. There isample evidence that the previousstandard was not doing an adequate jobof protecting workers from theserespiratory hazards, and that exposureto these hazards has continued to causeadverse health effects among exposedworkers. An analysis of OSHAinspection data from 1976 through 1982,when the previous standard had been ineffect for between five and eleven years(Ex. 33–5), found that in most cases(55.6%) where respirators were used toprotect employees from excessive levelsof air contaminants, respiratoryprotection programs were deficient inone or more elements, thus increasingthe potential for employee exposure.Even more significant was the fact thatin 72.1% of inspections in which anoverexposure to a substance listedunder 29 CFR 1910.1000 was cited,respirator use did not comply with therespiratory protection standard. OSHAperformed a similar analysis ofenforcement data for 1990–1996, andfound similar levels of noncompliance.[See also Work-Related Lung DiseaseSurveillance Report, 1994; USDHHS,CDC, DHHS (NIOSH) Number 94–120.]The provisions of the new respiratorstandard are designed to regulate howan employer selects, maintains, fit tests,and trains employees in the proper useof respiratory equipment, and to provideemployers with the tools needed toimplement an effective respiratoryprotection program. OSHA hasconcluded that the new standard willeliminate many of the unnecessaryillnesses and deaths described in thissection.

C. Responses to Advisory Committee onConstruction Safety and Health

The revised respirator standardreplaces the previous respiratoryprotection standard in the constructionindustry (29 CFR 1926.103). Since thisrevision affects the construction

industry, the September 1985preproposal draft standard waspresented to the Advisory Committeefor Construction Safety and Health(ACCSH) for its comments. The ACCSHcomments, combined with the othercomments received, were considered inpreparing a revision of the September1985 draft proposal.

As part of the Notice of ProposedRulemaking (NPRM) approval process,the revised NPRM was presented at theMarch 1987 ACCSH meeting and theCommittee’s comments were presentedto OSHA at the August 1987 meeting(Ex. 39). OSHA responded to theCommittee’s comments in the NPRM,published in November, 1994. As notedin that response, OSHA modified thedraft proposal to respond to theconcerns of the Committee (59 FR58931–58935).

The final standard replaces theprevious construction industry standardfor respiratory protection, 29 CFR1926.103, with an amended 29 CFR1926.103. The provisions of theprevious respiratory protection standard(29 CFR 1926.103) are deleted by thisaction. The title, Respiratory Protection,will remain in the Code of FederalRegulations but will now be followed bythe statement ‘‘Respiratory protectionfor construction employment is coveredby 29 CFR 1910.134.’’ The full text ofthis new standard will be printed in thegeneral industry standards, and theconstruction standard will reference therevised 29 CFR 1910.134.

The Agency’s responses to theCommittee’s specific concerns follow:

Paragraph (a)—Permissible PracticeThe Construction Advisory

Committee recommended thatparagraph (a)(1) of the standard bechanged to require that all feasibleengineering controls be used byemployers and that the employerdemonstrate that engineering controlsare not feasible before respirators maybe used. The recommended change alsowould have eliminated the requirementthat appropriate respirators be usedwhile engineering controls are beinginstalled. OSHA has stated elsewhere inthe summary and explanation section ofthis preamble that paragraph (a)(1) ofthe previous standard remainsunchanged in the new final standardbecause this paragraph was notproposed for revision and was thereforenot a subject of rulemaking in thisproceeding. The purpose of theRespiratory Protection standard is toimprove the level of protection providedto employees who use respirators toprotect them from respiratory hazards,regardless of whether that use occurs in

an environment where engineeringcontrols are in place.

The Committee proposed thatparagraph (a)(2) be modified to requirethat employers provide respirators toemployees exposed to contaminantconcentrations when the concentrationreaches one-half the PEL or TLV, andthat employees be required to wearthem before the PEL is exceeded. Toaccompany this revision the Committeeproposed a new definition establishingan ‘‘action level’’ of one-half the PEL forall regulated substances. OSHA has notadopted this ACCSH recommendationbecause the recommended changes arebeyond the scope of this rulemaking.

Paragraph (b)—DefinitionsACCSH suggested that OSHA add a

definition for ‘‘Grade D breathing air’’ tothe standard. The properties of Grade Dbreathing air are listed in paragraph (i)of the final standard, Supplied AirQuality and Use. OSHA believes thatrepeating these elements in thedefinition section is redundant andunnecessary.

The Committee also recommendedthat the rule include a definition for‘‘competent person,’’ as defined in 29CFR 1926.32(f). The competent personwould review the respiratory protectionprogram and perform the function of therespiratory program administratorrequired in paragraph (c)(2) of theproposal. OSHA has not included adefinition of competent person in thestandard because 29 CFR 1926.32(f)already has such a definition. OSHArecognizes, however, that, inconstruction settings, the competentperson is often also the administrator ofthe respirator program.

The Committee also recommendedthat the NIOSH RecommendedExposure Limits (RELs) be used alongwith the TLVs, to define a hazardousexposure level in the absence of a PEL.This point is no longer relevant becausethe concept of ‘‘hazardous exposurelevel’’ is not included in the finalrespiratory protection standard.

The proposal would have limited theuse of air-purifying respirators forhazardous chemicals with poor orinadequate warning properties. ACCSHrecommended a change to thedefinitions of ‘‘inadequate warningproperties’’ and that OSHA add a newdefinition for ‘‘odor threshold.’’ Becausethe final standard takes a differentapproach to determining when air-purifying respirators are appropriate,OSHA has not adopted the changesrecommended by ACCSH.

ACCSH also suggested that OSHArevise the proposed definition ofmaximum use concentration (MUC). In


the final standard the definition of MUChas been reserved, pending completionof a subsequent stage of this rulemakingthat will concentrate on establishingOSHA Assigned Protection Factors(APFs).

The Construction AdvisoryCommittee also recommended replacingthe proposal’s definition of ‘‘respirator;’’because the final standard contains nodefinition of ‘‘respirator,’’ thissuggestion has not been adopted. TheCommittee also recommened revisingthe proposed definition of ‘‘service life.’’However, since OSHA’s definition ofthis term has been broadened in thefinal rule and the rule contains detailedrequirements for change schedules forcartridges and canisters, ACCSH’sconcerns have largely been addressed.

Paragraph (c)—Respirator ProgramParagraph (c)(1) of the proposal

contained a requirement that theemployer establish a respirator programthat ‘‘covers’’ certain elements, asapplicable. OSHA has followed theCommitee’s recommendation thatOSHA change the word ‘‘cover’’ to‘‘include’’ but not removed the phrase‘‘as applicable,’’ as recommended by theCommittee, because not all elements ofthe program apply in all situations, andthus the ‘‘as applicable’’ language isappropriate.

The Committee also recommendedthat OSHA add an element to thewritten respirator program onprocedures for monitoring the workenvironment, using monitoring resultswhen selecting respirators, and selectingthe most protective respirators insituations where monitoring cannot beperformed (as is often the case inconstruction). OSHA considered thiscomment in drafting the final standard,which permits the employer to makereasonable estimates of exposure as partof the respirator selection process. Inmost cases, as discussed in the summaryand explanation of paragraph (d),monitoring results will form the basis ofa reasonable estimate. Where theemployer cannot estimate exposure, theatmosphere must be consideredimmediately dangerous to life or health(IDLH). For IDLH atmospheres, the mostprotective respirators are required.

One of the elements in the writtenrespirator program, paragraph (c)(1)(vi),states that the program shall includeprocedures to ensure proper air qualityfor atmosphere-supplying respirators.ACCSH asked OSHA to add the words‘‘quantity and flow’’ to provide moredirection for employers on what theprocedures should cover. OSHA agreesand has revised the wording of thiselement accordingly.

ACCSH recommended that OSHAsubstitute the term ‘‘competent person’’in paragraph (c)(2) for the language‘‘person qualified by appropriatetraining and/or experience.’’ Thisrecommendation has already beendiscussed above, in connection withACCSH’s comments on paragraph (b).

The written respiratory protectionprogram, in paragraph (c)(3), is requiredto reflect current workplace conditionsand respirator use. The Committeeurged OSHA to add the term ‘‘training’’to this element. OSHA has not done sobecause training is addressed in anotherprogram element. The Committee alsorecommended that OSHA add toparagraph (c) a provision allowingemployees and designatedrepresentatives access to exposure andmedical records maintained by theemployer. Because this requirement isalready included in 29 CFR 1910.1020,the medical and exposure records accessstandard, and referenced in this finalrespiratory protection standard, theAgency has not done so.

Proposed paragraph (c)(5) requiredemployers to make the written programavailable to affected employees,designated representatives, and OSHA.The Committee requested thatemployers be required to send a copy ofthe program to the OSHA SpecialAssistant for Construction. However, theproposed requirement has been movedto paragraph (m) of the final standard,which requires that all written materialsmaintained under the standard be madeavailable upon request to affectedemployees and the Assistant Secretary.This requirement should meet any needthat may arise for copies of the writtenprogram.

The Committee further recommendedthat the written respirator program bemaintained and made available toemployees at the job site, and that themedical and monitoring resultspertaining to respirator use be availableat the work site as well. The finalstandard in paragraph (m) now requiresemployers to allow employees toexamine and copy written programsupon request. Access to medical andmonitoring records for employeesexposed to toxic substances or harmfulphysical agents is regulated by OSHA ina separate standard, 29 CFR 1910.1020.That standard applies to constructionworkplaces as well as general industryworkplaces and requires the employerto ensure that access to medical andmonitoring records is provided in areasonable time, place, and manner(1910.1020(e)(1)(i)). Nothing in the finalrespiratory protection standard isintended to alter this requirement.

Paragraph (d)—Selection of Respirators

In its review of paragraph (d) of theproposal on selection of respirators, theCommittee requested OSHA to add anew provision that would requiremonitoring for contaminants when air-purifying respirators are used. Thisrequest is related to therecommendation for mandatorymonitoring, discussed above. The finalstandard requires that employers makereasonable estimates of employeeexposure levels when selecting allrespirators, not just air-purifying ones.Even if current monitoring results areunavailable, employers must base theirexposure estimates on reliable data,which might include, for example, theresults of past monitoring for similarconstruction jobs. Extensive discussionof this issue is contained in thesummary and explanation section ofthis preamble for paragraph (d). OSHAbelieves that allowing exposureestimates that may be based on pastmonitoring and other representativedata makes sense for the constructionindustry, where jobs are often short-lived and current monitoring datarelating to specific employees/operations may not be available whenrespirators must be selected. Becausethe final standard allows employers torely on reasonable estimates of exposureas well as monitoring results, OSHA hasnot added a requirement to the standardmandating that employers ‘‘obtain’’needed information, as recommendedby the Committee.

The Committee also recommendedremoval of the proposed phrase ‘‘whenthey exist’’ to modify the requirementthat employers select only NIOSH-approved respirators. Instead, theCommittee recommended use of themost protective respirator available, anSCBA or supplied air respirator, in caseswhere no approved air-purifyingrespirator exists. OSHA has removed thephrase ‘‘when they exist’’ from the finalstandard, for reasons explained in thesummary and explanation discussionrelating to paragraph (d).

The Committee urged OSHA toinclude poor odor warning properties asa reason for prohibiting the use of air-purifying respirators, and to removeproposed paragraph (d)(6)(ii), which,under limited circumstances, wouldhave allowed their use with substanceswith poor odor warning properties.Final paragraph (d)(3) modifies theproposal, and places many limitationson air-purifying respirator use withgases and vapors, regardless of theexistence of warning properties.

The Committee objected to the use ofair-purifying respirators in an


atmosphere with an oxygen content of19.5 percent at altitudes of 14,000 feetor below; in the Committee’s view,supplied air respirators should berequired in this situation. OSHAcontinues to treat atmospheres ataltitudes of 14,000 feet or below thathave oxygen concentrations of at least19.5% as non-oxygen-deficient, and torequire atmosphere-supplyingrespirators in these atmospheres.OSHA’s reasons for this determinationare detailed in the summary andexplanation section for paragraph (d).

Paragraph (e)—Medical EvaluationsThe Committee recommended that a

mandatory medical examination berequired in accordance with ANSIZ88.2, and that the standard include alist of diseases and conditions thatshould be considered in determining anindividual’s ability to wear a respirator.The final standard allows employers torely on a screening questionnaire toidentify employees with specifiedconditions that will require follow-upmedical examinations. Thequestionnaire specifies medicalconditions that OSHA has determinedoften relate to an employee’s ability touse a respirator. OSHA believes that thisprovision responds to the Committee’sconcern.

Based on the comments of ACCSHand others, OSHA has decided toeliminate the proposed exemption foremployees wearing respirators for nomore than 5 hours per week, for thereasons explained below in theSummary and Explanation. The finalrule also reflects the Committee’srecommendation that the medicalopinion provided to the employerinclude only limitations on theemployee’s ability to use a respirator.

The Committee recommended thatOSHA add a provision to this paragraphrequiring the employer to inform theperson performing the medicalexamination of the atmosphericcontaminants to which the employeewould be exposed. The final standardmeets this concern by requiring that thephysician or other licensed health careprofessional (PLHCP) receive a copy ofthe employer’s written respiratorprogram, and information about otherenvironmental conditions an employeemay encounter; this information willallow the medical professional to judgewhether the employee is medicallycapable of wearing the respirator.

The final rule allows an employerwho has, within the preceding 12months, provided his or her employeeswith a medical evaluation that fulfillsthe requirements of the revised standardto rely on the results of that evaluation.

OSHA believes that this provision isresponsive to the Committee’s concernthat limitations be placed on the‘‘portability’’ of medical evaluations.

The Committee recommended thatOSHA add a new provision to paragraph(e) to require that the employer providea powered air-purifying respirator oratmosphere-supplying respirator to anyemployee found medically unable towear a negative pressure respirator butotherwise able to perform the task to bedone. The final standard requires theemployer to provide a PAPR to anemployee when the PLHCP informs theemployer that the employee has amedical condition that may place theemployee’s health at increased risk ofmaterial impairment if the employeeuses a negative pressure respirator(paragraph (e)(6)(ii)) and is thusresponsive to the Committee’s concern.

Paragraph (f)—Fit TestingWith respect to fit testing procedures,

the Committee recommended thatproposed paragraph (f)(1) be rewritten tostate that respirators must fit theemployee so as to ensure that noexposure above the TLV or ceiling leveloccurs. OSHA agrees with theCommittee’s emphasis on fit testing andbelieves that the final rule’s fit testingrequirements and the fit test protocolsin an appendix to the standard willensure that employees are protectedfrom the overexposures of concern tothe Committee.

The Committee also suggestedclarifying that a fit test is requiredwhenever a different make or sizerespirator is used or when the facialcharacteristics of the employee change.The final rule addresses both of thesepoints.

The Committee recommendedlimiting the fit testing requirements totight-fitting negative pressurerespirators. This issue, and OSHA’sreasons for requiring fit testing of alltight-fitting respirators, is discussed inthe fit testing section of the Summaryand Explanation. OSHA has also deletedthe proposed provision, objected to bythe Committee, that would have allowedthe employer to use a qualitative fit testfor selecting respirators for employeeswho require fit factors greater than 10 insituations where outside contractorswho do the quantitative fit testing arenot available.

Paragraph (g)—Respirator UseParagraph (g)(1) of the final standard

adopts the proposed provisionprohibiting the use of respirators thatrely on a tight facepiece fit when facialconditions such as a beard or scarringwould prevent such fits. The Committee

urged OSHA to extend this provision tocover loose-fitting respirators as well astight-fitting ones. OSHA explains in theSummary and Explanation for thisparagraph that conditions such as abeard or facial scarring would have noeffect on the performance of loose-fittinghoods or helmets, and OSHA thereforedoes not regard it as appropriate tomake this change.

Employees who wear glasses wererequired in proposed paragraph (g)(4) towear them in a manner that does notinterfere with the facepiece seal of therespirator. The final standard continuesthis requirement (paragraph (g)(l)(ii)).The Committee suggested an additionalrequirement stating that, where theemployee must wear corrective lensesand the respirator requires that these beof special design, the employer providethe lenses at no cost to the employee.OSHA believes, however, that such arequirement is not necessary because, inmost cases where negative pressurerespirators may be worn, half-masks areacceptable, and half-masks eliminate theconcern about corrective glassesinterfering with facepiece seal. Becausethe final standard allows contact lensesto be worn, full facepiece respirators canbe worn by persons needing correctivelenses; contact lenses obviously do notinterfere with facepiece seal. Thus, thefinal rule gives employers severaloptions for addressing this concern ofthe Committee’s.

Paragraph (h)—Maintenance and Care ofRespirators

The Committee urged OSHA to addthe phrase ‘‘on paid time’’ to thisparagraph to ensure that employers notrequire employees to clean theirrespirators on their own time. OSHAhas decided in the final rule simply torequire employers to ensure thatrespirators are cleaned according tomandatory procedures or theirequivalents. OSHA believes that thisapproach is appropriate because therecord demonstrates that on-site,employer-supervised cleaning is theprevalent cleaning procedure and thestandard’s rigorous requirements forcleaning respirators will limit off-sitecleaning of respirators by employees.

Paragraph (k)—TrainingThe training section of the proposal

would have required that employersprovide a training program foremployees who are required to wearrespirators. The Committee urged OSHAto add language to paragraph (k)(1) torequire employers to provide, conductand document the effectiveness of thetraining program. The final standardtakes a more integrated approach in that


it requires employers to evaluate theentire respiratory protection programrather than the training programspecifically.

Paragraph (m)—Recordkeeping

OSHA has adopted the Committee’srecommendation to add the phrase ‘‘andmake available’’ to proposed paragraph(m)(1)(iii), which required employers tomaintain records of medical evaluationsin accordance with 29 CFR 1910.1020,the Access to Employee Exposure andMedical Records standard (seeparagraph (n)(1) of the final rule).

Appendix B—Recommended Practices

Appendix B–1 of the standardcontains practices for performingpositive and negative pressure facesealchecks. Respirator wearers are requiredby paragraph (g)(iii) to perform afaceseal check before entering the workarea either by following the mandatoryfaceseal check methods in Appendix B–1 or by following the respiratormanufacturer’s recommended method,if the employer shows that themanufacturer’s method is as effective asthe required methods. The Committeeurged OSHA to add new fit checkmethods to Appendix B–1, and OSHAhas responded to this recommendationby allowing the methods suggested bythe Committee if they are as effective asthe methods in the Appendix.

ACCSH also recommended thatOSHA issue a separate respiratorstandard for the construction industry.OSHA has reviewed the Committee’scomments to identify whichconstruction-specific concerns call forprovisions that differ from thoseapplicable to general industry. First,many of the final standard’s provisionsare stated in performance language,which is flexible enough toaccommodate differences in particularworkplaces or industries. For example,approved fit test systems, bothquantitative and qualitative, are portableand can be used on construction worksites as well as in fixed industrialfacilities. Another example is the finalrule’s requirement for medicalsurveillance; the frequency of medicalreevaluation is now event driven, whichwill greatly simplify evaluations foremployees who frequently changeemployment, as is the case with manyconstruction workers. Thus, OSHAbelieves that the final rule is responsiveto the Committee’s concerns about theuniqueness of the construction industryand is sufficiently flexible to be used onworksites in this sector.

D. Assigned Protection Factors

OSHA is reserving the sections of thisstandard addressing assigned protectionfactors (APFs) pending furtherrulemaking. OSHA is working diligentlyto complete the reserved portions of thestandard. In the interim, OSHA expectsemployers to take the best availableinformation into account in selectingrespirators. As it did under the previousstandard, OSHA itself will continue torefer to the NIOSH APFs in cases whereit has not made a differentdetermination in a substance-specificstandard.

E. Small Business Considerations

Pursuant to 5 U.S.C. 605(b) of theRegulatory Flexibility Act, OSHAcertified to the Small BusinessAdministration that the proposedrespiratory protection standard wouldnot have a significant impact on asubstantial number of small entities.

For the purposes of fulfilling therequirements of the RegulatoryFlexibility Act, the Agency in itsPreliminary Regulatory Impact Analysis(PRIA) [Ex. 57] examined the impact ofthe standard on a number of differentsmall establishment-size classes (1–7employees, 8–19 employees, etc).Although some economies of scaleassociated with the proposed standardwere noted, the Agency found that,given the modest costs perestablishment and the limited impact ofthe proposed regulatory revisions as awhole, the standard would not imposea significant economic impact on asubstantial number of small entities.These findings were summarized in theNPRM (59 FR 58894). At the time thatOSHA published the NPRM for thisrulemaking (Nov. 15, 1994), the Agencytransmitted the certification setting forththis conclusion, along with the fullPRIA, to the Small BusinessAdministration.

In developing the final standard, theAgency has conducted a screeninganalysis to identify any significantimpacts on a substantial number ofsmall entities. The details of thescreening analysis are presented in theFinal Economic Analysis, which isavailable in the docket; a summary ofthe analysis appears in section VI. Basedon this screening OSHA has againdetermined that the final rule will notimpose a significant impact on asubstantial number of small entities.The costs of the standard will equal nomore than 0.02 percent of revenues forsmall firms in any affected industry, andwill therefore pose no threat of businessdisruption, whether these costs areabsorbed by affected firms or passed on

to consumers. OSHA therefore certifiesthat the final rule will not have asignificant impact on a substantialnumber of small entities.

Nevertheless, the Agency hasdesigned the standard to minimizeimpacts on all affected establishments,and particularly on small entities.OSHA’s special consideration of smallbusinesses is in accord with theAgency’s continuing policy to remainsensitive to the needs of small entitiesaffected by Agency regulations.

Provisions that recognize the specialneeds of small businesses are discussedin more detail under specific sections ofthe Summary and Explanation of thestandard, Section VIII. Examples ofprovisions where consideration wasgiven to small businesses in makingregulatory decisions include:—Reduction in the number of repeat fit

tests required for quantitative fittesting;

—Allowing employers to use aquestionnaire (Appendix C is anexample) as a minimal medicalevaluation tool to ascertain anemployee’s ability to use respirators,rather than requiring a hands-onphysical examination;

—Allowing medical evaluations to beconducted either by a physician or byanother licensed health careprofessional (PLHCP), which willreduce medical surveillance costswithout compromising employeeprotection;

—Making the frequency of medicalevaluations, after the initialassessment, event-related instead oftime-related, e.g., only requiring suchevaluations when specific conditionsindicate a need for a reevaluation;

—Reducing the amount of paperworkrequired in connection with medicalevaluations. OSHA’s previousstandard required a physician todetermine pertinent health andphysical conditions, and furtherrequired that the respirator user’smedical status be reviewedperiodically (for instance, annually).Historically, employers have hadphysicians evaluate their employees’physical conditions, and havemaintained records documentingthose evaluations;

—Revising the requirements fordisinfecting respirators from ‘‘aftereach use’’ to ‘‘as necessary to bemaintained in a sanitary condition’’ toallow flexibility for small businesses;

—Requiring only that tags be used todocument respirator inspections,rather than requiring written records;and

—Allowing the employer to obtain acertificate of analysis of breathing gas


from the supplier rather thanrequiring employers to conduct gasanalyses themselves.In the Small Business

Administration’s Annual Report toCongress, a summary of SBA’scomments to the respirator docket (Ex.54–318) was provided. (Note that thesecomments pertain to the proposedrather than final rule.) SBA’s commentshave been examined alongside otherswith regard both to the proposal and itssupporting economic analysis. Asindicated, many of SBA’s suggestionshave been adopted; the SBA’s commentson the Preliminary Regulatory ImpactAnalysis are discussed in detail in theeconomic impact chapter of the FinalEconomic Analysis.

Revised 29 CFR 1910.134 is intendedto serve as a ‘‘building block’’ standardwith respect to future standards thatmay contain respiratory protectionrequirements; that is, future standardsthat regulate respirator use incontrolling employee exposure tohazardous conditions will refer toprovisions in the final respiratoryprotection standard. Further, OSHA hasfound that the respirator provisions ofexisting substance-specific standards(Asbestos, Cadmium, Lead, etc.) wereespecially in need of revision in view ofnewly revised § 1910.134. Except for alimited number of respirator provisionsunique to each substance-specificstandard, the remaining regulatory texton respirators now reads virtually thesame for each of these standards. Forexample, all provisions addressingrespirator use, selection, and fit testingwere deleted from the substance-specific standards, making thesestandards consistent with the finalrespiratory protection standard withrespect to these requirements. TheAgency believes that the revisions beingmade to 29 CFR 1910.134 aresufficiently comprehensive to allowdeletion of those provisions in thesubstance-specific standards thatduplicated provisions in the revisedfinal rule. A provision was retained onlywhen it addressed conditions (forexample, medical evaluation) that wereunique and/or integral to the substance-specific standard.

The Agency concludes that deletionof duplicative provisions from thesubstance-specific standards willenhance compliance, especially forsmall businesses, and will thus willimprove the protection afforded toemployees who use respirators.

IV. Certification/Approval ProceduresSection 1910.134(b)(8) of the previous

standard required that only thoserespirators approved jointly by NIOSH

and MSHA be used by the employer.The current respirator testing andapproval regulation, 30 CFR 11, whichauthorized the Bureau of Mines andNIOSH to jointly approve respiratoryprotection devices, was promulgated onMarch 25, 1972 at 37 FR 6244. OnNovember 5, 1974 the MineEnforcement Safety Administration(MESA) succeeded the Bureau of Minesand joined NIOSH in jointly approvingrespirators. Following the transfer ofMESA to the Department of Labor,where it became the Mine Safety andHealth Administration (MSHA),authority was transferred on March 24,1978 to MSHA for joint approval withNIOSH of respirators. Most of theBureau’s respiratory testing methods,developed in the 1950s or earlier, werechanged in the 1970s to reflect changesin testing technology.

NIOSH initiated revision of 30 CFR 11in 1980. A public meeting was held inJuly 1980 to address the certificationprogram. On August 27, 1987, NIOSHpublished a notice of proposedrulemaking (52 FR 32402) that wouldhave allowed NIOSH to certifyrespirators under the new 42 CFR part84 regulations, replacing the currentjoint NIOSH/MSHA 30 CFR 11certification regulations. The proposedNIOSH certification regulationscontained new and revisedrequirements for testing andcertification of respirators, and includeda set of assigned protection factors forvarious classes of respirators. Publichearings on the first draft of the NIOSHproposal were held in January 1988. Onthe basis of the comments received,NIOSH prepared a revised proposal forfurther public comment. On June 8,1995 NIOSH published revisedrespirator certification procedures forparticulate respirators (60 FR 30336)and recodified the previous certificationstandards for the other respirator classesas 42 CFR Part 84. These certificationprocedures address N, P and R classparticulate respirators at 95%, 99%, and99.7% levels of effectiveness.Additional public comment was soughtat public meetings convened in June1996 to assist NIOSH in preparation offuture rulemakings that will continuethe revision of the certificationprocedures for other classes ofrespirators. In October 1997, NIOSHannounced the intended priority orderfor these future rulemakings. Relevantaspects of these proceedings arediscussed in the Summary andExplanation.

V. Significance of RiskRespirators are used by American

workers as a means of protection against

a multitude of respiratory hazards thatinclude chemical, biological, andradiological agents. Situations in whichrespirators are relied upon to provideprotection from these hazards includethose that involve immediately life-threatening situations as well as routineoperations where engineering controlsand work practices are not able toprovide sufficient protection from thesehazards. In these situations, respiratorsmust ‘‘seal off’’ and isolate the worker’srespiratory system from thecontaminated environment. The riskthat a worker will experience an adversehealth outcome when relying onrespiratory protection is a function ofthe toxicity or hazardous nature of theair contaminants present, theconcentrations of the contaminants inthe air, the duration of exposure, andthe degree of isolation provided by therespirator. When respirators fail or donot provide the degree of protectionexpected by the user, the user is placedat an increased risk of any adversehealth effects that are associated withexposure to the respiratory hazardspresent. Therefore, it is critical thatrespirators perform as they are designedto do to ensure that users are not at anincreased risk of experiencing adverseeffects caused by exposure to respiratoryhazards.

OSHA has discussed the nature ofadverse health effects caused byexposure to airborne chemical hazardsmany times in previous rulemakingefforts (see, for example, Appendix A ofthe Hazard Communication standard, 29CFR 1910.1200 and the preambles toany of OSHA’s single substancestandards codified in 29 CFR 1910.1001to 1910.1052). In all instances whereOSHA has promulgated new or revisedPELs for chemical air contaminants,OSHA has determined that the healtheffects associated with exposure to thecontaminants represent materialimpairment of health because the effectsare life-threatening, cause permanentdamage, or significantly impair theworker’s ability to perform his or her jobin a safe manner. As discussed inSection VI of this preamble, OSHAexpects that thousands of illnesses andhundreds of fatalities that are presentlybeing caused by exposure to hazardoussubstances will be avoided annuallyamong respirator wearers as a result ofimprovements and clarifications madeto the earlier standard by this final rule.

Evidence on current workplaceexposure levels confirms that respiratorsare needed in many work situations toprotect workers against serious work-related illness. To illustrate, OSHAidentified several substances thatrepresent a range of adverse effects and


for which OSHA’s IntegratedManagement Information System (IMIS)database has documented workplaceexposures that exceed the current PELsfor these substances. The effectsrepresented by this subset of the IMISand the associated substances for whichthere are documented overexposuresinclude:—Sudden death/asphyxiation—carbon

monoxide, carbon dioxide;—Loss of lung function—wood dust,

welding fume, manganese fume,copper fume, cobalt metal fume,silica;

—Central nervous systemdisturbances—carbon monoxide,trichloroethylene;

—Cancer—chromic acid, wood dust,silica; and

—Cardiovascular effects—carbonmonoxide.When respirators are used during

operations where exposures exceedOSHA’s PEL, OSHA believes that thereis little or no margin that would protectthe worker in the event that therespirator does not perform as well asdesigned or expected. For all of thesubstances for which OSHA haspromulgated a comprehensive healthstandard (i.e., Arsenic, 29 CFR1910.1018; Asbestos, 29 CFR 1910.1001;Benzene, 29 CFR 1910.1028; Lead, 29CFR 1910.1025; Ethylene Oxide, 29 CFR1910.1047), OSHA has determined thatexposure above the PEL is associatedwith a significant risk of materialimpairment of health, and believes as amatter of policy that exposures belowthe PEL may be associated with risklevels that are significant. That is, thereis no exposure level near or somewhatabove the PEL that can be considered tobe at a low or insignificant risk level.Therefore, where workers perform jobsthat result in exposures above the PELfor any of these substances, use ofproperly functioning respirators isessential to ensure that workers are notplaced at significant risk of materialimpairment of health.

Throughout this preamble, OSHA hasdemonstrated that adequate fit testing,proper respirator selection, workertraining, and thorough inspection andmaintenance are essential elements of arespirator program. Without theserequirements, OSHA believes that thereis a greater chance that a respirator userwill inhale potentially dangerous aircontaminants, either by improperselection of equipment, excessiverespirator leakage, improper use of therespirator, or any combination of these.This section presents an analysisconducted by OSHA to evaluate theimproved protection to workers who use

respiratory protection equipment by thetype of effective respirator programrequired by the final rule.

In the context of a respiratoryprotection program, the health riskpresented to workers can be representedas the risk that a respirator will fail toprovide some minimum expected levelof protection, which increases thepossibility that the user of the respiratorwill be overexposed to a harmful aircontaminant. This presumes thatrespirators will be selected and used inwork settings where exposure toambient concentrations of aircontaminants poses an unacceptablehealth risk, and, if the respiratorperforms as expected, the wearer will beprotected from that risk. For example,an employer who provides a half-mask,chemical cartridge respirator foremployee use might typically assumethat the respirator will filter out 90percent of the contaminant and base hisor her choice of respirator on thatassumption. If the respirator performsless effectively than expected, theemployer’s expectation that therespirator will provide effectiveprotection will not be fulfilled.

This concept of risk differs from thatused by OSHA in its substance-specifichealth standards, in which the Agencytypically defines risk as the probabilitythat a worker will acquire a specificwork-related illness. Quantifying thatkind of risk requires the analysis of datathat relates the magnitude or intensity ofexposure to the incidence or prevalenceof adverse effects seen among exposedpopulations or experimental animals. Incontrast, the kinds of hazardoussituations covered by the finalrespiratory protection standard arevaried in terms of the nature of thehazard present (i.e., acute, chronic, orboth), the frequency and magnitude ofexposure, and the types of illnessesassociated with exposure to thosehazards. As a consequence, the healthrisks addressed by the final rule cannotbe described in terms of an illness-specific risk, but instead relate to themore general probability that arespirator will provide insufficientprotection causing the wearer to beexposed to a dangerous level of one ormore air contaminants.

Certain studies, referred to as‘‘workplace protection factor’’ (WPF)studies, have attempted to measure theeffectiveness of respirators under actualconditions of use in the workplace. TheWPF is a measure of the reduction inexposure achieved by using respiratoryprotection and is represented by anestimate of the ratio of the concentrationof a contaminant found in theworkplace air to the concentration

found inside the respirator facepiecewhile the respirator is being worn. Asthe degree of protection afforded by therespirator increases, the WPF increases.Alternatively, the degree of protectionprovided by a respirator can beexpressed as a penetration value, whichis the reciprocal of the WPF and reflectsthe ratio of the concentration ofcontaminant inside the facepiece to theconcentration outside. For example, aWPF of 50 equates to a penetrationvalue of 0.02 and means that theconcentration inside the respiratorfacepiece is one-fiftieth of the ambientlevel.

Because WPF studies are designed toevaluate the field effectiveness ofrespiratory protection equipment, studyprotocols usually have been designed tominimize factors that can reducerespirator performance. Such factorsinclude selecting the wrong type ofrespirator for the working conditionsunder which the study is beingconducted, use of poorly fittingrespirator facepieces (i.e., testing ofrespirator fit is routinely done in well-conducted WPF studies), inadequatetraining of wearers in proper respiratoradjustment and use, or excessiveleakage caused by malfunctioning ordirty respirator parts. Typically, WPFstudy protocols include procedures forproperly selecting respirators andensuring that they are in good workingorder, assigning respirators to workerson the basis of valid qualitative orquantitative fit tests, training wearers onhow to adjust strap tension properly anduse the respirator, and ensuring thatneither facial hair nor other personalprotective equipment is likely tointerfere with respirator fit. In addition,workers included in WPF studies areusually monitored throughout theperiod that respirators are worn to verifythat the equipment is being properlyused. All of these conditions reflect theprincipal elements of a strong respiratorprogram in which respiratorperformance is optimized; therefore, theresults from a good WPF study canmirror the results obtained by anemployer who implements a well-runrespiratory protection program.

To quantitatively evaluate the impactof implementing a good respiratorprogram on respirator performance,OSHA identified several WPF studiesthat were conducted using methods thatreflect a comprehensive program, andcompared these results to otherworkplace studies that did not employall of the elements of a good program.Quantitative approaches are used todevelop (1) aggregate estimates ofrespirator effectiveness in both thepresence and absence of a good


respiratory protection program, and (2)estimates of the frequency with whichworkers are likely to achieve inadequateprotection while using a respirator,given the presence or absence of a goodunderlying program. All of the studiesused in this analysis pertain to theeffectiveness of half-mask, negative-pressure respirators, and all arecontained in OSHA’s rulemaking docket(H–049).

Many of the well monitored WPFstudies conducted were reviewed byNelson et al. in 1995 (Ex. 64–514); theseauthors selected data from seven suchstudies to evaluate the overall fieldeffectiveness of half-mask, negative-pressure respirators. Each of the studiesdescribed by Nelson et al. ensuredselection of properly fitted respiratorseither by an accepted qualitative fit test(QLFT) (i.e., isoamyl acetate orsaccharin) or by a quantitative fit test(QNFT) where only respirators thatprovided a minimum protection factorto the wearer of at least 100 wereselected. Each of these studies providedfor worker instruction in properrespirator use, and workers weremonitored during each study to ensureproper use. An additional six studieswere reviewed by Nelson et al. but wererejected either because they allegedlyused biased sampling methods todetermine ambient and in-facepiececontaminant concentrations or becausethe authors believed that improper orinvalidated fit test procedures wereemployed.

In the studies selected by Nelson et al.for analysis, workers used elastomericor disposable respirators equipped withdust-mist, dust-mist-fume, or high-efficiency particulate (HEPA) filters, andthe collection of studies represented arange of workplace exposure situations,including pigment production, metalsrefining, asbestos exposure duringbrake-repair work, welding, and spraypainting. Geometric Mean (GM) WPFvalues from these studies ranged from47 to 3,360, with an overall GM WPF of290. The 5th percentile WPF from thedata set was estimated to be 13, with a95% confidence interval of 10–18.Nelson et al. concluded from theanalysis of the overall data set that theassigned protection factor of 10 for half-mask, negative-pressure respirators wasreasonable given that a WPF of less than10 would not likely occur more than 5percent of the time. In addition, Nelsonet al. found no significant difference inthe field performance of disposablerespirators compared to elastomericmodels. OSHA has not conducted adetailed comparative evaluation of WPFvalues obtained from disposable vs.elastomeric respirators; if, in fact,

disposable respirators provide lessprotection than elastomeric respirators,the WPFs that can be achieved under agood respirator program will beoverstated in this analysis since Nelsonet al.’s compiled data reflect the use ofboth types of respirators.

Each of the studies reviewed byNelson involved worker exposures todusts. OSHA could identify only oneWPF study, by Galvin et al. in 1990 (Ex.64–22), that examined respiratoreffectiveness against exposure to avapor-phase contaminant rather than aparticulate. In this study, WPFmeasurements were taken on a group of13 styrene workers who used half-mask,air-purifying respirators equipped withchemical cartridge filters. All employeeswere assigned respirators based onpassing an irritant smoke fit test, and allwere trained on how to properly don therespirator and conduct fit checks. In-mask and ambient styreneconcentrations were measured over one-hour periods, during which employeeswere instructed not to readjust thefacepiece. Chemical cartridges werechanged with each new sampling periodto ensure that there was nobreakthrough. In-mask styreneconcentrations were adjusted upwardsby 40 percent to account for pulmonaryretention, which avoided potentiallyoverestimating the WPF. The GM WPFfor the overall cohort was reported to be79, with a geometric standard deviation(GSD) of 3.51. There was no significantdifference in WPF values between thoseworkers engaged in relatively physicaloperations, such as spraying, comparedto those performing less physical worktasks. The GM WPF found by Galvin etal. for styrene-exposed workers lieswithin the range of GM WPF valuesreported in the studies reviewed byNelson for worker cohorts exposed toparticulate-contaminated environments.

Nelson in his 1995 report (Ex. 64–514)excluded the Galvin et al. study fromhis analysis because fit tests wereperformed using the irritant smokeprotocol. As discussed in the Summaryand Explanation section of thispreamble, OSHA has determined thatthe irritant smoke qualitative fit testprovides a valid, effective test ofrespirator facepiece fit. The proceduresused by Galvin et al. to ensure adequateworker training and respirator use areconsistent with the elements of apermissible respirator program, andOSHA, therefore, finds it appropriate toinclude this study in the set of WPFstudies that are representative ofeffective respiratory program practices.

In contrast, OSHA has identified threestudies where investigators alsodetermined WPF values for half-mask,

negative-pressure respirators, but wherefew steps were taken to ensuremaximum respirator performance.OSHA believes that these studiesillustrate the relative lack of protectionafforded by respirators when certaincritical elements of the respiratoryprotection program are missing orinadequate. The studies identified byOSHA are those by Toney and Barnhartin 1972 (Ex. 64–68), Moore and Smithin 1976 (Ex. 64–49), and Harris et al. in1974 (Ex. 27–11).

Toney and Barnhart (Ex. 64–68)conducted a WPF study to evaluate theeffectiveness of half-mask, chemical-cartridge respirators on reducingexposures of spray painters to solventvapors and aerosols. Data were obtainedfrom painters working at 39 differentsites and included both in-mask andambient concentrations. WPFs werefound to be low; from the raw datapresented in the study, OSHAcalculated a GM WPF of 3.8 for solventexposure (GSD=2.28, N=39) and a GMWPF of 11.4 for aerosol exposure(GSD=4.12, N=40). Penetration testsperformed on unused respiratorcartridges of the same types used in thefield indicated that the poor WPFsachieved in the field tests were causedby poor respirator fit and a lack ofrespirator maintenance, and were notdue to any inherent defect in thecartridges. The authors concluded thatrespirators being used by painters werenot effective and cited several reasons,all pointing to the lack of a respiratoryprotection program at the facilitiestested. For example, 28 percent ofrespirators used by the painters werepoorly maintained. Some of theconditions found by the investigatorsincluded deteriorating rubber on thefacepieces, the presence of stuck orwarped valves, missing head straps, andevidence of leakage around the cartridgeseal. In addition, it was apparent thatsome of the cartridges had not beenchanged for extended periods of time.Many of the facilities studied suppliednon-approved respiratory protectivedevices (respirators were approved bythe Bureau of Mines at the time of thestudy), and most had no formal trainingor maintenance program in place. Theauthors found that ‘‘* * * managementand workers are extremely uninformedon the subject of selection, use, and careof respiratory protective devices.’’ (Ex.64–68, p. 93).

The second study, conducted byMoore and Smith in 1976 (Ex. 64–49),measured WPF values obtained byworkers exposed to sulfur dioxide (SO2)during a furnace charging operation at acopper smelter. Three models of half-mask, chemical cartridge respirators


were tested on each of nine workers; in-mask and ambient SO2 concentrationswere measured during the furnacecharging operation while the respiratorswere worn. There is no indication in thestudy that qualitative or quantitative fittesting was performed to verify adequatefacepiece fit. A total of 81 samples werecollected, 5 of which were excludedfrom the analysis because the subjectsremoved or lifted the respiratorfacepiece during the sampling period.Average ambient SO2 concentrationsvaried in the range of 53 to 61 mg/m3

(20.4 to 23.5 ppm) during the samplingperiod. Geometric mean WPF valuesreported for each of the three models ofrespirator were 22.1 (SD=22.6), 18.4(SD=14.2), and 12.9 (SD = 11.0). Mooreand Smith concluded that the overallprotection afforded by the respiratorswas poor, and that between one-thirdand one-half of the protection factorsachieved would be below 10, theaccepted minimum protection factor forthat type of respirator. Reasons given bythe authors for the poor fits observedamong the subject workers included thepossibility that strap tension was notproperly adjusted (the authors did notcontrol or monitor strap tension),variation in facial hair (despite the lackof beards or wide sideburns), andnormal work activities that caused headmotion and deep breathing associatedwith heavy work.

The third study is that of Harris et al.in 1974 (Ex. 27–11), who evaluated theperformance of five half-mask dustrespirators among 37 miners working in4 coal mines. In-mask and ambient dustmeasurements were made throughoutthe workshifts, during which minersintermittently used respiratoryprotection. Thus, this study differs fromthe others described above in that theratio of in-mask to outsideconcentrations included periods of timewhere the respirator was not worn, incontrast to the typical WPF study. Theratio of in-mask to outside concentrationdetermined during periods ofintermittent respirator use, termed the‘‘effective protection factor’’ (EPF), isnot directly comparable to WPF valuesbecause, to the extent that workersspend time in contaminatedatmospheres without respiratory

protection, the WPF will tend tounderstate the actual protectionobtained while the respirator is beingworn. However, according toPoppendorf in 1995 (Ex. 54–512), it ispossible to use EPF data to estimate theWPF that was likely to have beenachieved during periods of respiratoruse if both of the following are knownor can be estimated: (1) The fraction oftime during which the respirator wasnot worn by the subject, and (2) the ratioof contaminant concentration in areaswhere the respirator was worn to that inareas where the respirator was not worn.Poppendorf (Ex. 54–512) described themathematical relationship between theEPF and WPF and suggested that thelikely range of average WPF valuesachieved by the miners during periodsof respirator use was 3.6 to 5.7. Thisestimate of WPF is based on anobservation by Harris et al. that minerswore their respirators about half of thetime during the sampling periods, andan assumption by Poppendorf (Ex. 54–512) that the dust levels in the air whilerespirators were worn were at least 5times higher than airborne dust levelsduring periods of respirator non-use.OSHA believes that the latterassumption is reasonable given thatHarris et al. reported that, for the mostpart, miners wore their respirators onlywhen visible airborne dust was present.Harris et al. noted that the hard hatsworn by the miners interfered withproper respirator strap positioning andadjustment; OSHA believes that thisfactor, as well as the apparent lack of fittesting, is likely to have contributed tothe low protection factors experiencedby the miners.

OSHA believes that the studiesdescribed above demonstrate thatimproved respirator performance can beachieved under actual workplaceconditions if fit testing is used to selectrespirators, if respirators are clean andin good working order, and if employeesare properly trained and supervised intheir use. This is evident when thesummary statistics from aggregateprotection factor data obtained fromfield studies on groups of employeesusing respirators in the absence of astrong respirator program (i.e., Mooreand Smith, Toney and Barnhart, Harris

et al.) are compared with those obtainedfrom cohorts using respirators under thecondition of a strong program (i.e., thestudies reviewed by Nelson and thestudy by Galvin et al.). Summaryprotection factor data from these studiesare presented in Table V–1 as geometricmean and mean WPF values, and thegeometric standard deviation (GSD) ofthe distribution of WPF values. Fromthese summary statistics, OSHAcomputed a weighted geometric meanWPF across cohorts exposed toparticulate contaminants to compare thecentral tendency in protection factorsachieved both with and without anadequate underlying respirator program(see footnote on Table V–1).

In general, groups of employees usingrespirators against particulate exposuresunder a strong program achieved anoverall GM protection factor about 25-fold higher than groups usingrespirators without the elements of astrong respiratory protection program.In studies that did not implement all ofthese elements, mean WPF valuesamong the particulate-exposed workercohorts tested ranged from about 6 to22. Mean WPF values for particulate-exposed worker cohorts included in theWPF studies where elements of a goodprogram were implemented ranged from72 to 2,400, with the mean WPF fromone study estimated to be 11,500. Theresults from studies that examinedrespirator effectiveness against gas orvapor, also included in Table V–1, showan 8-fold difference in overall GM WPFvalues. With only one exception, the 95percent confidence intervals around theGM WPF values computed from thestudies reflecting inadequate programpractices do not overlap with thosecomputed from the studies reflectingstrong program elements (see Table V–1); thus, the hypothesis that there are nodifferences in the GM WPF valuesbetween the two groups of studies isrejected. This analysis suggests thatimplementation of a good respiratoryprotection program containing theelements described by the final rule cancontribute to a substantial increase inthe overall performance of respiratorsused in actual workplace settings, asmeasured by the mean WPF acrossgroups of workers.


TABLE V–1.—SUMMARY RESULTS FROM WORKPLACE PROTECTION FACTOR (WPF) STUDIES AND ESTIMATED FRE-QUENCIES OF RESPIRATOR FAILURE, BASED ON A ONE-FACTOR ANOVA ANALYSIS OF DATA FROM WORKPLACE PRO-TECTION FACTOR (WPF) STUDIES

Study Geometric meanWPF (95% C.I.1)

Geometricstandarddeviation

MeanWPF

Estimated percent of workers with:

MeanWPF ≤10 2

MeanWPF ≤2 2

WPF ≤10at least

5% of thetime 3

WPF ≤2 atleast 5%

of thetime 3

Studies Reflecting Inadequate Program ElementsParticulate Exposure

Toney and Barnhart [1972] (Ex. 64–68) ..... 4 11.4 (3.2–39.6) 4 4.12 31.1 76.8 9.0 100 60.4Harris et al. [1974] (Ex. 27–11)

Low Estimate ....................................... 5 3.6 (1–17.9) 5 2–93 6.4 99.7 38.8 100 96.4High Estimate ....................................... 5 5.7 (1.6–20.4) 5 2.93 10.2 97.0 12.5 100 82.3

Weighted Geometric Mean .............. 6 5.6Gas/Vapor Exposure

Moore and Smith [1976] (Ex. 64–69)Respirator A ......................................... 15.29 (8.3–28.1) 7 2.36 22.1 36.2 <0.01 98.9 1.9Respirator B ......................................... 13.72 (7.7–24.4) 7 2.15 18.4 41.3 <0.01 99.7 0.5Respirator C ......................................... 9.59 (4.8–19.2) 7 2.16 12.9 83.1 <0.01 100 9.0

Toney and Barnhart [1972] (Ex. 64–68) ..... 4 3.8 (1.2–11.9) 4 2.28 5.3 100 14.7 100 95.7Weighted Geometric Mean .............. 6 9.4

Studies Reflecting Good Program ElementsParticulate Exposure

Dixon and Nelson [1984] 8 .......................... 3360 (3101–3640) 4.8 11,498 <0.01 <0.01 <0.01 <0.01Gaboury and Burd [1989] 8 ......................... 47 (31–72) 2.5 72 0.2 <0.01 30.1 <0.01Lenhart and Campbell [1984] 8 ................... 166 (120–228) 3.8 405 0.1 <0.01 9.0 0.02Nelson and Dixon [1985] 8 .......................... 258 (192–347) 5.2 1004 0.7 <0.01 14.5 0.3Gosselink et al. [1986] 8 .............................. 96 (75–123) 2.3 136 <0.01 <0.01 0.1 <0.01Colton and Mullins [1992] 8 ......................... 147 (117–185) 2.5 224 <0.01 <0.01 0.1 <0.01Myers [1990] 8 ............................................. 346 (256–468) 7.2 2,428 2.8 0.1 22.2 1.7

Weighted Geometric Mean .............. 6 142Gas/Vapor Exposure

Galvin et al. [1990] (Ex. 64–22) .................. 79 (54–115) 3.5 173 1.1 <0.01 31.7 0.2

1 95% confidence interval of the geometric mean WPF calculated as follows for simultaneous confidence intervals: y±SD÷√nɽ tn¥1,1¥α/2,α=1¥(1¥0.05)1/N

where n is the number of WPF measurements in each study and N is the number of studies being compared (i.e., 10 for particulate studiesand 5 for gas/vapor studies).

2 Calculated from equation 9 as described in the text; δ = 0.1 for WPF = 10, δ = 0.5 for WPF = 2.3 Calculated from equation 10 as described in the text; κ = 0.1 for WPF = 10, κ = 0.5 for WPF = 2.4 Calculated by OSHA from raw data presented by the authors.5 Range of WPF values estimated by Popendorf [1995] (Ex. 54–512), from effective protection factor values (EPF) reported by Harris et al.

GSDs calculated by OSHA from median and mean EPF values reported by Harris et al.6 Calculated as a weighted geometric mean as follows: exp[(∑lnGM/(lnGSD)2)/∑(1/(lnGSD)2)].7 Calculated by OSHA from median and mean WPF values reported by Moore and Smith.8 Studies reviewed by Nelson [1995] (Ex. 64–514).

The three WPF studies representingdeficient program practices were allconducted 10 to 20 years earlier thanthe WPF studies reflecting goodprogram elements. Thus, differencesbetween the two groups of studies inworking conditions, processes andexposures, or respirator equipment andtechnology could confound thecomparison of respirator effectivenessmeasures. OSHA is not aware of anyrecent studies that have been conductedthat were designed to evaluate theimpact of respirator program elementson respirator effectiveness, nor arerecent studies available that haveattempted to measure respiratoreffectiveness under conditions of a poorrespiratory protection program. OSHAbelieves that this analysis of programimpacts on respirator performance isbased on the best available data.However, OSHA has consideredwhether confounding factors related to

the elements of a good respiratorprogram may also have contributed tothe differences in respiratorperformance reported by the two groupsof WPF studies. For example, respiratorfit can be adversely affected by vigorouswork activity requiring head motion anddeep breathing. Heavy work loads alsocontribute to respirator discomfort,which may cause a worker to wear arespirator too loosely. The nature of theair contaminant affects respiratorperformance in that different types ofrespirator filters have differentcapabilities in purifying contaminatedair and gas-phase contaminants andsmall-particulate aerosols pass morereadily through leak points than doaerosols comprised mostly of largerparticles.

OSHA does not believe that anysystematic differences in workingconditions or respirator technologycontribute substantially to the

differences in respirator effectivenessfound between the two groups of studiesincluded in the analysis. For example,both groups of studies represent a rangeof workplace situations that involvestrenuous and non-strenuous work. Inthe studies that do not reflect goodprogram practices, workers wereengaged in active, strenuous work(smelter operations and coal mining) aswell as less active work (spray painting).Similarly, studies that reflect goodprogram practices have also beenconducted on worker cohorts engaged inboth active work (metals refining) andless active work (spray painting, brakerepair). Both groups of studies alsoinvolve a range of contaminants,including both gas-phase and variouskinds of particulate. Some of the studiesreviewed by Nelson includedinformation on the size distribution of


particulates to which workers wereexposed, with the range across thesestudies including both respirable andnon-respirable particles. Other studiesincluded in the Nelson analysisreported that workers were exposed toboth dust and fume. Therefore, thedifferences in WPFs found between thetwo groups of studies cannot beexplained by differences in particulatesizes or characteristics. Both groups ofstudies also represent a variety of half-mask respirator designs and filters,including single-use respirators andrespirators equipped with dust/mist(i.e., non-HEPA) filters. OSHA believesit unlikely that the 14-fold difference inoverall WPFs between the two groups ofstudies can be primarily attributed toany fundamental differences inrespirator equipment or technology.Therefore, OSHA finds that thedifferences in WPF values obtainedfrom the two groups of studies are morelikely to reflect differences in how wellthe respirators fit the subject workers,the condition of the respiratoryequipment used, and the extent towhich the equipment was usedproperly, rather than any confoundingcaused by systematic differences inwork settings, the nature of theexposures, or the age of the WPFstudies.

The kinds of summary statisticspresented in Table V–1 have been usedby several investigators to demonstratehow poorly or how well respirators canprotect workers under actual conditionsof use (see, for example, Moore andSmith (Ex. 64–69), Nelson et al. (Ex. 64–

514)). However, such descriptivemeasures can only provide informationon the aggregate frequency distributionof protection factor values in a group ofworkers. Although it is useful to rely onsummary statistics from aggregateprotection factor data to make generalstatements about the effectiveness ofrespirators, such measures do notadequately convey information on thenumber or proportion of workers whoremain at risk of overexposure to aircontaminants despite the use ofrespiratory protection, or howfrequently an individual worker mightexperience poor fits.

Nicas (Ex. 156) and Nicas and Spearin 1992 (Ex. 64–425) have suggested thatusing statistics from aggregateprotection factor data does notadequately describe the true risk ofoverexposure to workers usingrespirators because the approach fails torecognize that there are two differentsources of variability that account forthe overall variation in protection factorvalues measured from a given cohort ofworkers. One source of variability inprotection factors is the variationtypically experienced by a single workerfrom one day to the next; this is termedwithin-worker variability. The secondsource of variability reflects theobservation that different workerswithin a group will achieve differentaverage protection factors over a givenperiod of time; this is termed between-worker variability. In a peer-reviewedarticle, Nicas and Spear (Ex. 64–425)have described a statistical model thataccounts for both sources of variability.

This model has been used by OSHA toestimate the following from theprotection factor studies describedabove to better characterize risks toworkers who use respirators both in theabsence of and under a strongrespiratory protection program:

(1) The proportion of workers who fail toachieve a long-term average protection factorat or above some specified target level,exposing the worker to an increased risk ofa chronic health hazard (i.e., a health hazardthat is typically associated with long-termcumulative exposure); and

(2) The proportion of workers who achievea protection factor below some specifiedtarget level at least 5 percent of the time thatthe respirator is worn, thus increasing thefrequency with which a worker may beexposed above an effect concentrationassociated with an acute health hazard.

The Nicas and Spear model (Exs. 64–425, 156) used by OSHA in this analysisis a one-factor analysis of variance andis described briefly as follows. Let Pdenote a penetration value experiencedby the wearer of a respirator during arandomly selected wearing time (P isdefined as the reciprocal of theprotection factor PF measured in theworkplace, or 1/PF). For example, a Pvalue of 0.1 for a respirator wearerreflects that a protection factor of 10 wasachieved in the workplace for thatindividual. If one were to measure thepenetration values among members of agroup of workers over time andaggregate the results, the totaldistribution of P values can be describedby the following parameters:

( )

( ) [ ] [ ] [ ]

( ) [ ] exp ln [ ] ln [ ]

1

2

3 2 2

P B W

GM P GM B GM W

GSD P GSD B GSD W

p

p

= × ×= × ×

= +

µµ

Where:P = the penetration value for a worker

for a particular wearing period,µp = the arithmetic mean penetration

value for the population,B = a lognormally distributed factor that

transforms µp to the arithmeticmean penetration value for theindividual worker, and

W = a lognormally distributed factorthat transforms µp × B to the P value

experienced by the individualworker for a particular wearingtime.

The factors W and B describe within-worker variability and between-workervariability, respectively.

Since workplace protection factorstudies typically report the geometricmean and geometric standard deviationof protection factor values obtained

from a cohort of respirator wearers (i.e.,GM[P] and GSD[P]), the parametersdescribed above for within-worker andbetween worker variability can beestimated as follows if the relationshipbetween GSD[B] and GSD[W] are knownor assumed. Let R represent the ratio ofGSD[W]/GSD[B]; then GSD[B] can beestimated from GSD[P] and R by therelationship

( ) [ ] exp ln ln ln ( ) ln .41

2

1

222 2 2GSD B R R GSD P R= −

+ + −( )


GSD[W], GM[B], and GM[W] areestimated by:

( ) [ ] [ ]

( ) [ ] /exp . ln ,

( ) [ ] /exp . ln .

5

6 1 0 5

7 1 0 5

2

2

GSD W GSD B R

GM W GSD W

GM B GSD B

= ∗= ∗ [ ]( )(

= ∗ [ ]( )(and

The arithmetic mean of the totaldistribution of penetration values acrossthe whole cohort, µp, is estimated by:

( )[ ]

( [ ] [ ])8 µ p

GM P

GM B GM W=

∗

Nicas (Ex. 156) defines two additionalvalues, δ and κ, that are based on theparameters described above. The valueδ represents the 95th percentile of thebetween-wearer distribution of averagepenetration values among a cohort of

respirator wearers; thus, there is a 5percent chance that a respirator wearerin the cohort could have an averagepenetration value of δ or higher. If δ isset to some penetration value reflectingsome minimum acceptable value of

protection, the probability that arespirator wearer would fail, on average,to achieve the minimum acceptablepenetration value is Pr(Z>z), where

( )ln ln ln [ ]

ln [ ]9 z

GM B

GSD B

p=

− +( )( )δ µ

and Z is the standard normal deviate. Byestimating the parameters µp, GM[B],and GSD[B] from WPF data, one canestimate the probability that a respiratorwearer could have an averagepenetration value greater than somespecified value δ.

The value κ is defined by Nicas (Ex.154) based on the distribution of eachworker’s 95th percentile P value andrepresented the P value experienced atleast 5 percent of the time by 95 percentof workers in the cohort. If κ is set tosome minimum acceptable P value, the

estimated probability that a respiratorwearer could fail to achieve theminimum P value at least 5% of thetime is Pr(Z>z), where

( )ln ln ln [ ] . ln [ ] . ln [ ]

ln [ ]10

1 645 0 5 2

zGM B GSD W GSD W

GSD B

p=

− + + ( ) − ( )[ ]κ µ

and Z is the standard normal deviate.Thus, the proportion of workers whofail to achieve a P value of κ at least 5percent of the time can be determinedby estimating the parameters µp, GM[B],and GSD[W] from WPF data.

The following hypothetical exampleillustrates OSHA’s use of the model toestimate the risk to workers ofexperiencing an overexposure whileusing respiratory protection. Supposethat the WPF values obtained from agroup of workers using half-mask,negative-pressure respirators are foundto have a geometric mean of 50 (i.e.,GM[P] = 1/50 = 0.02) and a geometricstandard deviation of 3.0 (GSD[P] = 3.0).Furthermore, from one of the WPFstudies reviewed by OSHA (Galvin etal.) (Ex. 64–22), it was reported thatwithin-worker variability exceededbetween-worker variability in workplace

protection factors, with the ratioGSD[W]/GSD[B] = 1.5. From equations 4through 7 above, and assuming that R =1.5, then GSD[B] = 1.73, GSD[W] = 2.60,GM[W] = 0.63, and GM[B] = 0.86. Thearithmetic average of the cohort’s Pvalues, µp, is estimated from equation 8to be 0.037. If a protection factor of lessthan 10 (the NIOSH minimum assignedPF for half-mask respirators) isconsidered to place the worker at risk ofan overexposure, then equation 9predicts a probability of 1.8 percent thata worker in the group would beexpected to have an average WPF valueof 10 or less (i.e., δ is set to 0.1 inequation 9); that is, 1.8 percent of thegroup of respirator wearers wouldfrequently encounter situations wherethey are working in a hazardousenvironment without the minimumprotection expected from the respirators

being used. By equation 10, there is asubstantial probability (47 percent) thata worker in the cohort would notachieve a minimum protection factor of10 at least 5 percent of the time thatrespirators are used (i.e., κ is set to 0.1in equation 10).

OSHA used the Nicas and Spearmodel, the summary data from the WPFstudies reviewed above, and the methodoutlined in the example described aboveto estimate the probability that arespirator wearer would fail to receiveadequate protection from theirrespirator; the detailed results of thisanalysis appear in Table V–1, andsummary findings are listed in Table V–2. From the studies that reflect the lackof an adequate respiratory protectionprogram, the Nicas and Spear modelpredicts a high probability (between 36and 100 percent) that a wearer would


not achieve an average protection factorof 10. Data from two of these studies byToney and Barnhart (Ex. 64–68), andHarris et al. (Ex. 27–11), when used inthe model, suggest a probability ofbetween 13 and 39 percent that theaverage WPF for a respirator wearercould be 2 or less, which may beconsidered equivalent to receiving nolong-term protection at all. In contrast,workers included in the studies

reflecting good respirator programelements would be expected toexperience low WPFs much lessfrequently. The probability that a wearerwould attain an average WPF of 10 orless is estimated to be between <0.01and 3 percent. Results from the studiesthat reflect good respiratory programpractices also indicate that long-termaverage WPF values at or below 2 wouldrarely occur. The results from this

analysis demonstrate that deficienciesin implementing a good respiratorprogram can greatly increase the chancethat the wearer of a negative-pressurerespirator will receive less than theminimum expected average protectionfrom the respirator over the long-term,thus increasing the chance that theworker will be exposed to a higherchronic health risk.

TABLE V–2.—SUMMARY ESTIMATES OF THE PROBABILITY OF ACHIEVING INADEQUATE FITS FOR HALF-MASK, NEGATIVE-PRESSURE RESPIRATORS UNDER DEFICIENT AND GOOD RESPIRATORY PROTECTION PROGRAMS

Quality of respirator program

Percent probability that wearer willachieve

Average work-place fit factor of

less than 10

Workplace fit fac-tor of less than 10at least 5 percentof time that res-pirator is worn

Deficient ....................................................................................................................................................... 36–100 99–100Good ............................................................................................................................................................. <0.01–3 <0.01–32

OSHA’s analysis (Tables V–1 and V–2) also demonstrates that workers usingrespiratory protection under a deficientprogram will be exposed morefrequently to higher concentrations ofairborne contaminants, which mayincrease the risk that the worker willexperience acute health effects. TheNicas and Spear model applied to thestudies that reflect inadequate respiratorprograms predicts nearly a 100 percentchance that a protection factor of lessthan or equal to 10 would beexperienced at least 5 percent of thetime. Under conditions of a goodrespirator program, use of the modelsuggests no more than a 32 percentchance that WPFs of less than or equalto 10 will occur more than 5 percent ofthe time.

OSHA finds that, without an adequaterespiratory protection program in place,a substantial fraction of respirator usersare at risk of being overexposed tohazardous air contaminants due to poorrespirator performance. The studiesconducted under conditions of a poorrespirator program, when analyzedusing the Nicas and Spear model,suggest a greater than 50 percentprobability that the wearer of a half-mask, negative-pressure respirator willregularly fail to attain the expectedminimum level of protection, and thatthe chance of receiving essentially noprotection is substantial. OSHAconsiders these risks of overexposure tobe significant. The studies reviewed byNelson and the Galvin study indicatethat these risks are considerably lowerin situations where respirators are usedin conjunction with the implementation

of strong respiratory protection programelements such as appropriate fit testing,adequate employee training, use ofclean respirators in good working order,and regular monitoring of employees toensure proper respirator use. Thus,OSHA finds that implementation of acomprehensive respiratory protectionprogram, such as the one prescribed bythe final rule, will substantially reducethe risk of overexposure that is due torespirator failure. Because suchoverexposures can place workers at asignificant risk of health impairment, asdescribed earlier in this section, OSHAalso finds that promulgation of the finalrule will substantially reduce thesignificant health risks associated withthose overexposures.

VI. Summary of the Final EconomicAnalysis

In the Final Economic Analysis,OSHA addresses the significant issuesrelated to technological and economicfeasibility and small business impactsraised in the rulemaking process. Thisanalysis also explains in detail theAgency’s findings and conclusionsconcerning pre-standard (baseline)conditions, such as respirator programpractices, in establishments in theregulated community, and discusseshow and why the requirements of thestandard are expected to reduceemployee exposures. The preamble tothe revised rule and the Final EconomicAnalysis are integrally related andtogether present the fullest statement ofOSHA’s reasoning concerning thisstandard. The Final Economic Analysis

has been placed in the rulemakingdocket.

This analysis of OSHA’s revisedRespiratory Protection standard (29 CFR1910.134) has been conducted inaccordance with Executive Orders (EOs)12866 and 12875, the RegulatoryFlexibility Act (as amended in 1996),the Small Business RegulatoryEnforcement Fairness Act (SBREFA),the Unfunded Mandates Reform Act(UMRA) and the Occupational Safetyand Health Act. The standard is a‘‘significant’’ rule as defined by EO12866, a ‘‘major’’ rule as defined by Sec.804 of SBREFA, and a ‘‘significant’’ ruleas defined by UMRA.

The purposes of this Final EconomicAnalysis are to:

• Describe the need for a revisedstandard governing the use ofrespirators;

• Identify the establishments,industries and employees potentiallyaffected by the standard;

• Evaluate the costs, benefits,economic impacts and small businessimpacts of the standard on affectedfirms;

• Assess the technological andeconomic feasibility of the standard foraffected establishments, industries, andsmall businesses; and

• Identify the availability of effectivenon-regulatory and alternativeregulatory approaches.

OSHA’s final Respiratory Protectionstandard covers the use of respiratoryprotection in general industry,construction and shipyard employment,as well as marine terminals andlongshoring. In all, about 5 million


1 Approximately 5% of these respirator-usingemployees would be subject to OSHA’s substance-

specific health standards rather than to thisstandard.

employees are estimated to userespirators. 1 Workers use respirators toprotect themselves from a wide varietyof occupational exposures. Respiratorsare used, at least to some extent, invirtually every industry, although theextent of respirator use varies byindustry. Manufacturing andconstruction have relatively heavyrespirator use; in contrast, use in manyservice industries is very limited.

Chapter II of the economic analysisdescribes the pattern of respirator usewithin each affected industry. Todevelop this profile, the Agencyanalyzed the results of several OSHA-sponsored nationwide surveys. Theresults of OSHA’s analysis appear inTable VI–1. The Agency estimates thatapproximately five percent of workerswear respirators at some time, and thatapproximately 1.3 million

establishments, or about 20 percent ofall establishments, have employees whouse respirators. Approximately 900,000of these establishments are very small,i.e., have fewer than 20 employees. Fora discussion of the number of firmsidentified by the Small BusinessAdministration (SBA) as small, seeChapter V.

TABLE VI–1.—NUMBER OF RESPIRATOR USERS AND THEIR EMPLOYERS BY INDUSTRY

SIC and industry Total employ-ment

Number ofrespiratorwearers

Total numberof establish-

ments

Number of es-tablishments

with respiratorwearers

07 Agricultural services .................................................................................. 555,686 48,262 95,956 25,46408 Forestry ..................................................................................................... 17,716 2,764 2,251 95013 Oil and gas extraction ............................................................................... 257,694 46,180 18,502 3,31315 General contractors and operative builders ............................................. 1,096,289 202,284 180,998 70,83516 Heavy construction, except building ......................................................... 679,578 99,668 34,332 13,40317 Special trade contractors .......................................................................... 2,731,774 491,928 382,528 115,38020 Food and kindred products ....................................................................... 1,498,078 87,589 21,049 8,89921 Tobacco products ..................................................................................... 37,189 2,022 119 4722 Textile mill products .................................................................................. 615,683 66,989 6,245 1,93723 Apparel and other textile products ............................................................ 972,060 26,431 24,293 5,23824 Lumber and wood products ...................................................................... 675,081 89,970 37,087 15,92225 Furniture and fixtures ................................................................................ 476,488 56,141 11,515 7,67526 Paper and allied products ......................................................................... 627,746 41,313 6,478 2,61627 Printing and publishing ............................................................................. 1,500,580 19,185 65,416 6,39328 Chemicals and allied products .................................................................. 851,720 230,405 12,371 10,74429 Petroleum and coal products .................................................................... 112,984 29,647 2,117 1,39830 Rubbber and miscellaneous plastics products ......................................... 915,166 53,800 16,048 6,80531 Leather and leather products .................................................................... 104,747 4,406 2,025 32432 Stone, clay, and glass products ............................................................... 471,639 69,904 16,208 8,79833 Primary metal industries ........................................................................... 655,556 133,012 6,726 4,10534 Fabricated metal products ........................................................................ 1,371,072 124,289 36,416 17,13435 Industrial machinery and equipment ......................................................... 1,749,735 96,161 54,436 25,54536 Electronic and other electronic equipment ............................................... 1,424,351 65,930 17,073 6,89537 Transportation equipment ......................................................................... 1,601,554 185,783 11,420 7,64938 Instruments and related products ............................................................. 878,379 35,188 11,419 4,20739 Miscellaneous manufacturing industries ................................................... 375,501 22,751 17,183 6,79340 Railroad transportation .............................................................................. 49,200 1,790 1,000 22541 Local and interurban passenger transit .................................................... 366,657 13,337 18,603 4,19442 Trucking and warehousing ........................................................................ 1,633,543 59,497 115,531 26,04944 Water transportation ................................................................................. 162,478 7,458 8,412 60545 Transportation by air ................................................................................. 344,822 12,543 11,436 82246 Pipelines, except natural gas .................................................................... 17,143 2,808 811 52147 Transportation services ............................................................................. 363,103 22,428 47,858 3,44148 Communication ......................................................................................... 1,299,658 15,176 40,399 3,45749 Electric, gas, and sanitary services .......................................................... 924,373 187,298 21,040 10,14850 Wholesale trade—durable goods ............................................................. 3,414,441 373,644 317,418 118,38751 Wholesale trade—nondurable goods ....................................................... 2,504,260 289,619 185,908 70,19652 Building materials and garden supplies .................................................... 696,228 95,688 69,965 19,82253 General merchandise stores ..................................................................... 2,141,964 21,420 35,646 3,56554 Food stores ............................................................................................... 3,027,828 30,278 181,850 18,18555 Automotive dealers and service stations .................................................. 1,992,774 245,662 198,905 80,12156 Apparel and accessory stores .................................................................. 1,194,121 15,788 143,526 14,35357 Furniture and homefurnishings stores ...................................................... 754,024 12,348 112,254 11,22558 Eating and drinking places ....................................................................... 6,727,618 67,276 441,512 44,15159 Miscellaneous retail .................................................................................. 2,422,923 38,734 352,129 35,21360 Depository institutions ............................................................................... 2,095,049 20,950 102,622 10,26261 Nondepository institutions ......................................................................... 483,133 4,831 41,869 4,18762 Security and commodity brokers .............................................................. 449,826 4,498 34,325 3,43363 Insurance carriers ..................................................................................... 1,570,356 15,704 43,784 4,37864 Insurance agents, brokers, and service ................................................... 656,007 13,452 122,292 12,22965 Real estate ................................................................................................ 1,335,048 25,846 234,961 23,49667 Holding and other investment offices ....................................................... 254,172 3,016 27,420 2,74270 Hotels and other lodging places ............................................................... 1,527,126 15,271 52,874 5,287


2 OSHA believes that, for the purposes of thisrulemaking, the most reasonable way to summarizethe uncertainties in benefits estimates via a singlenumerical estimate is to use the expected value;that is, the average of all plausible values weighted

by their relative probabilities. For simplicity’s sake,OSHA will refer to this point estimate as the ‘‘bestestimate.’’

3 Because this regulation will not directly affectthe benefits for the estimated 5% of employees whowear respirators as a result of OSHA’s substance-specific health standards (except to the extent thatuniformity of provisions improve compliance), andthese respirator-wearing employees are included inthe benefits estimates presented here, the benefitsof the revised respiratory protection standard aresomewhat overestimated. In particular, deaths andillnesses caused by exposures to such OSHA-regulated substances as asbestos and lead may infact account for a disproportionate share (more than5%) of the occupational illnesses and deathsattributed by this analysis to the respiratorstandard. This means that OSHA’s benefitsestimates are likely to be overstated by more than5%. Nevertheless, OSHA believes that thesubstantial majority of the benefits resulting fromappropriate respirator use can be properlyattributed to the respirator standard.

4 Because this regulation does not directly affectthe costs for the estimated 5% of employees who

wear respirators as a result of OSHA’s substance-specific health standards, and these respirator usersare included in the cost estimates, the costs aresomewhat overestimated. Because costs areapproximately proportional to the number ofemployees affected, the magnitude of thisoverestimate is likely to be about 5%.

TABLE VI–1.—NUMBER OF RESPIRATOR USERS AND THEIR EMPLOYERS BY INDUSTRY—Continued

SIC and industry Total employ-ment

Number ofrespiratorwearers

Total numberof establish-

ments

Number of es-tablishments

with respiratorwearers

72 Personal services ...................................................................................... 1,252,777 45,854 200,520 23,84873 Business services ..................................................................................... 5,832,261 255,034 322,668 38,37575 Auto repair, services, and parking ............................................................ 903,806 110,528 174,635 70,34576 Miscellaneous repair services ................................................................... 439,495 5,103 72,763 3,81078 Motion pictures .......................................................................................... 500,889 5,009 42,457 4,24679 Amusements and recreation services ...................................................... 1,201,248 12,012 88,077 8,80880 Health services ......................................................................................... 10,403,118 217,118 471,873 108,33781 Legal services ........................................................................................... 962,374 17,417 158,335 15,83482 Educational services ................................................................................. 1,967,024 19,670 42,867 4,28783 Social services .......................................................................................... 2,028,694 20,287 145,998 14,60084 Museums, botanical, zoological gardens .................................................. 73,874 739 3,607 36186 Membership organizations ........................................................................ 2,062,501 26,275 238,868 23,88787 Engineering and management services ................................................... 2,589,839 27,483 249,846 24,98589 Services, n.e.c. ......................................................................................... 84,960 1,607 14,606 1,46192 Fire Departments (State Plan States) ...................................................... 126,500 126,500 9,283 9,283

Other public sector (State Plan States) .................................................... 7,677,000 114,570 203,158 20,316

Total ....................................................................................................... 98,768,281 4,953,568 6,494,122 1,281,945

Sources: DOL, OSHA Office of Regulatory Analysis; County Business Patterns, 1993; OSHA’s respirator, PEL, PPE, and Construction PELsurveys.

The new standard is programmatic innature, reflects current practice at manyfacilities, and does not require the useof new technology. Thus, OSHA findsthat the standard is clearlytechnologically feasible for affectedfirms of all sizes.

The benefits that will accrue torespirator users and their employers aresubstantial and take a number of forms.Chapter IV of the analysis describesthese benefits, both in quantitative andqualitative forms. The standard willbenefit workers by reducing theirexposures to respiratory hazards.Improved respirator selectionprocedures, better fit test procedures,and improved training, all areasstrengthened by the revised standard,will contribute substantially to greaterworker protection. Estimates of thebenefits of the standard are complicatedby uncertainties about the effectivenessof the standard and the number ofcovered work-related illnesses. TheAgency estimates that the standard willavert between 843 and 9,282 work-related injuries and illnesses annually,with a best estimate (expected value) 2 of

4,046 averted illnesses and injuriesannually. In addition, the standard isestimated to prevent between 351 and1,626 deaths annually from cancer andmany other chronic diseases, includingcardiovascular disease, with a bestestimate (expected value) of 932 averteddeaths from these causes.3

The annual costs employers in theaffected establishments are estimated toincur to comply with the revisedrespirator standard total $111 million.4

These costs, which are presented indetail in Chapter III of the full economicanalysis, are annualized over a 10-yearhorizon at a discount rate of 7 percent;Table VI–2 shows annualized costs byprovision of the standard. The mostcostly provisions are those requiringannual fit testing of respirators andannual refresher training. These twoprovisions together account forapproximately 90 percent of thestandard’s compliance costs. As a rule,costs are largely determined by theextensiveness of respirator use inaffected establishments. This analysisdid not attempt to factor in the offsettingvalue of cost savings from regulatorychanges, such as dropping the existingstandard’s prohibition against contactlens use, providing for greateruniformity for substance-specific healthstandard respirator provisions, orallowing employers to use licensedhealth care providers in addition tophysicians to perform medicalevaluations.


TABLE VI–2.—ANNUAL COST OF RESPIRATOR STANDARD REVISIONS FOR RESPIRATOR-USING ESTABLISHMENTS, BYPROVISION

SIC and industryRevisionwrittenplans

Annual fittesting

Annualtraining

Certifi-cation for

emergencyrespirator

inspections

Labelingfor sorbent

bedchanges

Record-keeping Total

07 Agricultural services .......................................... $31,755 $441,836 $298,047 $0 $0 $35,858 $807,49708 Forestry ............................................................. 1,228 25,475 13,849 0 0 2,054 42,60613 Oil and gas extraction ....................................... 8,769 734,048 315,180 41,551 0 34,312 1,133,86015 General contractors and operative builders ...... 141,534 2,992,402 1,909,631 0 479 150,297 5,194,34216 Heavy construction, except building ................. 32,027 1,534,132 736,976 0 2,109 74,053 2,379,29717 Special trade contractors .................................. 256,681 7,820,459 4,340,977 0 1,344 365,502 12,784,96320 Food and kindred products ............................... 21,109 1,006,778 428,004 86,371 0 65,078 1,607,33921 Tobacco products .............................................. 210 37,254 16,252 0 0 1,502 55,21822 Textile mill products .......................................... 4,349 728,823 286,222 9,703 0 49,773 1,078,87023 Apparel and other textile products .................... 7,864 226,658 101,380 0 0 19,638 355,54024 Lumber and wood products .............................. 27,997 972,293 489,510 16,750 0 66,848 1,573,39725 Furniture and fixtures ........................................ 13,119 623,774 289,781 53,627 0 41,712 1,022,01326 Paper and allied products ................................. 8,373 877,037 280,715 66,279 105 30,696 1,263,20527 Printing and publishing ...................................... 15,217 221,275 139,295 0 0 14,255 390,04128 Chemicals and allied products .......................... 33,159 4,194,240 1,656,678 741,170 763 171,191 6,797,20129 Petroleum and coal products ............................ 4,699 646,431 277,684 108,927 16 22,028 1,059,78530 Rubber and miscellaneous plastics products ... 14,100 676,734 284,187 2,068 0 39,974 1,017,06331 Leather and leather products ............................ 456 37,208 15,800 1,502 0 3,274 58,23932 Stone, clay, and glass products ........................ 20,743 1,018,192 464,833 28,365 11 51,939 1,584,08333 Primary metal industries .................................... 14,028 2,263,416 951,396 44,664 28 98,828 3,372,36034 Fabricated metal products ................................. 41,510 1,663,770 765,562 178,892 0 92,346 2,742,08135 Industrial machinery and equipment ................. 64,626 1,498,968 786,251 0 868 71,447 2,422,16136 Electronic and other electronic equipment ........ 17,103 917,414 388,929 24,483 657 48,986 1,397,57237 Transportation equipment ................................. 23,876 3,413,486 1,568,463 100,401 8,775 138,037 5,253,03838 Instruments and related products ..................... 10,299 516,278 230,813 1,626 333 26,145 785,49339 Miscellaneous manufacturing industries ........... 12,007 250,490 136,104 0 176 16,904 415,68240 Railroad transportation ...................................... 937 37,818 16,134 0 0 1,330 56,21941 Local and interurban passenger transit ............ 9,002 167,510 86,710 0 0 9,910 273,13142 Trucking and warehousing ................................ 64,666 791,301 511,259 570 0 44,206 1,412,00344 Water transportation .......................................... 1,588 136,318 65,312 0 0 5,541 208,76045 Transportation by air ......................................... 2,015 199,061 85,196 0 0 9,320 295,59246 Pipelines, except natural gas ............................ 1,637 87,121 31,182 0 15 2,086 122,04147 Transportation services ..................................... 6,150 256,532 135,948 0 0 16,664 415,29448 Communication .................................................. 9,141 282,097 141,518 0 0 11,276 444,03249 Electric, gas, and sanitary services .................. 32,542 3,736,483 1,662,243 359,209 4,581 139,162 5,934,22050 Wholesale trade—durable goods ...................... 241,074 5,545,911 2,737,719 6,687 0 277,618 8,809,00851 Wholesale trade—nondurable goods ................ 134,760 3,979,336 1,728,752 126,854 0 215,187 6,184,88852 Building materials and garden supplies ............ 24,193 922,814 418,187 0 0 71,096 1,436,29153 General merchandise stores ............................. 5,369 135,056 56,819 0 0 15,915 213,16054 Food stores ....................................................... 27,336 208,820 154,036 0 0 22,497 412,68955 Automotive dealers and service stations .......... 112,276 1,920,333 1,281,723 0 0 182,527 3,496,85856 Apparel and accessory stores ........................... 19,022 91,801 92,713 0 0 11,730 215,26657 Furniture and homefurnishings stores .............. 20,225 111,532 106,953 0 0 9,175 247,88458 Eating and drinking places ................................ 47,123 257,557 214,860 0 0 49,986 569,52659 Miscellaneous retail ........................................... 53,098 275,565 269,808 0 0 28,780 627,25060 Depository institutions ....................................... 20,271 207,313 135,320 0 0 15,566 378,47061 Nondepository institutions ................................. 10,608 51,626 53,951 0 0 3,590 119,77662 Security and commodity brokers ....................... 10,508 64,998 58,550 0 0 3,342 137,39763 Insurance carriers .............................................. 13,360 226,063 123,889 0 0 11,668 374,97964 Insurance agents, brokers, and service ............ 36,394 200,209 199,277 0 0 9,995 445,87565 Real estate ........................................................ 70,079 348,877 368,891 0 0 19,203 807,05167 Holding and other investment offices ................ 8,272 43,583 43,970 0 0 2,241 98,06670 Hotels and other lodging places ....................... 8,119 101,853 57,381 0 0 11,347 178,69972 Personal services .............................................. 26,015 552,641 270,488 0 0 34,069 883,21473 Business services .............................................. 58,974 3,325,952 1,172,726 0 0 189,490 4,747,14275 Auto repair, services, and parking .................... 93,387 970,308 881,030 0 0 82,122 2,026,84676 Miscellaneous repair services ........................... 5,735 61,214 54,759 0 0 3,791 125,49978 Motion pictures .................................................. 11,425 62,923 61,091 0 0 3,722 139,16079 Amusement and recreation services ................. 14,128 93,683 76,484 0 0 8,925 193,22080 Health services .................................................. 183,206 2,510,780 1,948,071 0 0 161,319 4,803,37681 Legal services ................................................... 47,661 253,320 256,703 0 0 12,941 570,62582 Educational services ......................................... 10,933 259,816 125,365 0 0 14,615 410,72983 Social services .................................................. 23,601 166,510 130,949 0 0 15,073 336,13384 Museums, botanical, zoological gardens .......... 891 8,995 6,036 0 0 549 16,47186 Membership organizations ................................ 57,115 316,483 304,939 0 0 19,523 698,06087 Engineering and management services ............ 74,480 380,740 390,356 0 0 20,420 865,99789 Services, n.e.c. .................................................. 4,082 28,754 22,201 0 0 1,194 56,231


TABLE VI–2.—ANNUAL COST OF RESPIRATOR STANDARD REVISIONS FOR RESPIRATOR-USING ESTABLISHMENTS, BYPROVISION—Continued

SIC and industryRevisionwrittenplans

Annual fittesting

Annualtraining

Certifi-cation for

emergencyrespirator

inspections

Labelingfor sorbent

bedchanges

Record-keeping Total

92 Fire Departments ............................................... 24,723 2,265,377 1,005,792 0 0 93,990 3,389,882Other public sector ............................................ 48,361 49,739 1,147,899 0 0 85,126 1,331,125

Total .......................................................................... 2,501,319 67,033,593 35,865,707 1,999,699 20,259 3,680,501 111,101,079

Source: Department of Labor, Safety and Health Administration, Office of Regulatory Analysis.

Chapter V of the economic analysisanalyzes the impact of these compliancecosts on establishments in affectedindustries. The standard is clearlyeconomically feasible: the cost in theaverage affected establishment is 0.002

percent of sales and 0.03 percent ofprofits; in the most heavily impactedindustry—business services, SIC 73—annualized compliance costs amount toonly 0.1 percent of estimated sales and1.22 percent of profits. In the next most

heavily impacted industry—SpecialTrade Contractors, SIC 17—costsamount only to 0.02 percent of sales and0.46 percent of profits. These results areshown in Table VI–3.

TABLE VI–3.—ANNUAL COST OF FINAL RESPIRATORY PROTECTION STANDARD AS A PERCENT OF SALES AND PROFITSOF RESPIRATOR-USING ESTABLISHMENTS

SIC and industry

Averagecompliancecost/estab-

lishment

Averagesales/estab-

lishment

Average prof-it/establish-

ment

Compliancecost as apercent of

sales


profits

07 Agricultural services ...................................................................... $32 $269,290 17,425 0.01 0.1808 Forestry ......................................................................................... 45 897,908 69,720 0.00 0.0613 Oil and gas extraction ................................................................... 364 11,234,630 1,021,330 0.00 0.0415 General contractors and operative builders .................................. 73 1,131,765 52,585 0.01 0.1416 Heavy construction, except building ............................................. 178 2,709,660 146,028 0.01 0.1217 Special trade contractors .............................................................. 111 476,348 24,098 0.02 0.4620 Food and kindred products ........................................................... 192 20,620,629 999,788 0.00 0.0221 Tobacco products .......................................................................... 1,169 869,935,367 204,319,114 0.00 0.0022 Textile mill products ...................................................................... 578 7,611,245 438,223 0.01 0.1323 Apparel and other textile products ................................................ 68 3,228,588 194,177 0.00 0.0324 Lumber and wood products .......................................................... 99 2,539,729 146,588 0.00 0.0725 Furniture and fixtures .................................................................... 140 3,571,798 216,729 0.00 0.0626 Paper and allied products ............................................................. 551 22,478,383 1,260,152 0.00 0.0427 Printing and publishing .................................................................. 61 2,096,632 152,975 0.00 0.0428 Chemicals and allied products ...................................................... 909 29,454,052 2,231,368 0.00 0.0429 Petroleum and coal products ........................................................ 1,053 143,210,471 6,292,581 0.00 0.0230 Rubber and miscellaneous plastics products ............................... 150 8,202,235 584,099 0.00 0.0331 Leather and leather products ........................................................ 187 7,267,252 429,429 0.00 0.0432 Stone, clay, and glass products .................................................... 183 4,184,931 228,219 0.00 0.0833 Primary metal industries ............................................................... 864 18,123,180 1,015,996 0.00 0.0834 Fabricated metal products ............................................................ 170 4,348,383 266,070 0.00 0.0635 Industrial machinery and equipment ............................................. 95 6,924,099 482,589 0.00 0.0236 Electronic and other electronic equipment ................................... 207 11,591,397 684,946 0.00 0.0337 Transportation equipment ............................................................. 724 44,334,058 1,948,012 0.00 0.0438 Instruments and related products ................................................. 187 10,720,444 763,426 0.00 0.0239 Miscellaneous manufacturing industries ....................................... 61 1,568,937 111,245 0.00 0.0640 Railroad transportation .................................................................. 249 NA NA NA NA41 Local and interurban passenger transit ........................................ 65 1,014,732 43,699 0.01 0.1542 Trucking and warehousing ............................................................ 54 1,286,872 58,437 0.00 0.0944 Water transportation ...................................................................... 345 NA NA NA NA45 Transportation by air ..................................................................... 359 3,106,975 197,717 0.01 0.1846 Pipelines, except natural gas ........................................................ 234 13,802,633 585,566 0.00 0.0447 Transportation services ................................................................. 121 23,585,180 8,076,137 0.00 0.0048 Communication ............................................................................. 128 1,894,095 82,755 0.01 0.1649 Electric, gas, and sanitary services .............................................. 677 15,622,527 2,485,402 0.00 0.0350 Wholesale trade—durable goods .................................................. 74 14,371,043 1,350,007 0.00 0.0151 Wholesale trade—nondurable goods ............................................ 89 2,282,652 102,134 0.00 0.0952 Building materials and garden supplies ........................................ 72 4,447,849 172,734 0.00 0.0453 General merchandise stores ......................................................... 60 1,075,912 36,708 0.01 0.1654 Food stores ................................................................................... 23 8,648,964 471,762 0.00 0.0055 Automotive dealers and service stations ...................................... 44 2,179,673 61,031 0.00 0.0756 Apparel and accessory stores ...................................................... 15 2,010,075 47,296 0.00 0.0357 Furniture and homefurnishings stores .......................................... 22 737,603 47,246 0.00 0.0558 Eating and drinking places ............................................................ 13 672,234 34,798 0.00 0.04


5 The Agency also examined the impact of thecosts of compliance on governmental entities

serving communities with fewer than 50,000people, and also found small impacts.

TABLE VI–3.—ANNUAL COST OF FINAL RESPIRATORY PROTECTION STANDARD AS A PERCENT OF SALES AND PROFITSOF RESPIRATOR-USING ESTABLISHMENTS—Continued

SIC and industry

Averagecompliancecost/estab-

lishment

Averagesales/estab-

lishment

Average prof-it/establish-

ment


sales


profits

59 Miscellaneous retail ....................................................................... 18 734,358 34,558 0.00 0.0560 Depository institutions ................................................................... 37 547,141 30,254 0.01 0.1261 Nondepository institutions ............................................................. 29 8,651,403 NA 0.00 NA62 Security and commodity brokers .................................................. 40 9,094,686 1,419,322 0.00 0.0063 Insurance carriers ......................................................................... 86 6,131,429 631,723 0.00 0.0164 Insurance agents, brokers, and service ........................................ 36 65,412,387 NA 0.00 NA65 Real estate .................................................................................... 34 674,913 NA 0.01 NA67 Holding and other investment offices ........................................... 36 500,929 46,869 0.01 0.0870 Hotels and other lodging places ................................................... 34 5,183,873 573,368 0.00 0.0172 Personal services .......................................................................... 37 1,243,240 97,027 0.00 0.0473 Business services ......................................................................... 124 128,952 10,164 0.10 1.2275 Auto repair, services, and parking ................................................ 29 975,693 74,455 0.00 0.0476 Miscellaneous repair services ....................................................... 33 358,494 22,775 0.01 0.1478 Motion pictures .............................................................................. 33 181,478 11,743 0.02 0.2879 Amusement and recreation services ............................................ 22 1,597,336 142,792 0.00 0.0280 Health services .............................................................................. 44 631,398 31,198 0.01 0.1481 Legal services ............................................................................... 36 1,167,682 71,435 0.00 0.0582 Educational services ..................................................................... 96 421,539 67,758 0.02 0.1483 Social services .............................................................................. 23 2,613,764 174,383 0.00 0.0184 Museums, botanical, zoological gardens ...................................... 46 351,713 16,137 0.01 0.2886 Membership organizations ............................................................ 29 560,217 40,331 0.01 0.0787 Engineering and management services ....................................... 35 320,236 15,070 0.01 0.2389 Services, n.e.c. .............................................................................. 38 1,030,962 81,876 0.00 0.0592 Fire Departments .......................................................................... 365 NA NA NA NA

other public sector ..................................................................... 66 NA NA NA NA

Source: Department of Labor, Occupational Safety and Health Administration, Office of Regulatory Analysis.

In the Preliminary Regulatory ImpactAnalysis developed in support ofOSHA’s 1994 Respiratory Protectionproposal [Ex. 57], the Agency examinedthe impact of the proposal on differentsizes of establishments. Based on thatanalysis, the Agency certified that theproposed standard would not have asignificant economic impact on asubstantial number of small entities.Upon review of comments and otherdata submitted to the record of thisrulemaking, the Agency has analyzedthe final rule’s impact on small entities,as defined by the Small BusinessAdministration (SBA) and inaccordance with the RegulatoryFlexibility Act. In addition, in order toensure that even the smallest entities arenot significantly impacted, the Agency

performed an analysis of impacts on thesmallest establishments, i.e., those withfewer than 20 employees.

The impacts of the standard on salesand profits did not exceed 1 percent forsmall firms in any covered industry,whether the analysis used the SBA’sdefinitions or the fewer-than-20-employee size class definition. Becausethe incremental costs of the final ruleare primarily related to the number ofrespirator users per establishment andbecause small entities do not have ahigher percentage of respirator usersthan large establishments, the standarddoes not have a differential impact onsmall entities. If the costs of compliancewere influenced by economies of scale,such effects would have beendemonstrated by OSHA’s analysis of the

smallest firms, i.e., those with fewerthan 20 employees. However, no sucheffects were seen, even among firms inthis smallest size-class. Therefore, theAgency has no reason to believe thatestablishments or firms in intermediatesize groupings, i.e., those in the rangebetween 20 employees and theemployment size cutoff for theapplicable SBA definition, wouldexperience larger impacts. Finding this,the Agency certifies that the finalRespiratory Protection standard will nothave a significant adverse economicimpact on a substantial number of smallentities. The results of OSHA’s analysisof small business impacts on firms 5

within the SBA’s size classifications areshown in Table VI–4.

TABLE VI–4.—ANNUAL COST OF THE RESPIRATORY PROTECTION STANDARD AS A PERCENT OF SALES FOR RESPIRATOR-USING SMALL FIRMS 1

SIC and industry Small business defi-nition 1

Numberof af-fectedfirms

Averagecompli-ance

cost perfirm

Averagesales per firm

Compli-ance

cost asa per-cent ofsales

Average prof-it per firm

Compli-ance

cost asa per-cent ofprofits

07 Agricultural services ........................................... $5 million 2 ............... 23,313 $36 $223,567 0.02 $14,466 0.2508 Forestry .............................................................. $5 million ................. 860 41 470,247 0.01 36,513 0.1113 Oil and gas extraction ........................................ 500 employees ....... 2,565 222 2,017,392 0.00 226,361 0.10


TABLE VI–4.—ANNUAL COST OF THE RESPIRATORY PROTECTION STANDARD AS A PERCENT OF SALES FOR RESPIRATOR-USING SMALL FIRMS 1—Continued

SIC and industry Small business defi-nition 1

Numberof af-fectedfirms

Averagecompli-ance

cost perfirm

Averagesales per firm

Compli-ance

cost asa per-cent ofsales

Average prof-it per firm

Compli-ance

cost asa per-cent ofprofits

15 General contractors and operative builders ...... $17 million ............... 70,232 75 954,486 0.01 43,794 0.1716 Heavy construction, except building .................. $17 million ............... 12,628 135 1,611,092 0.00 72,025 0.1917 Special trade contractors ................................... $7 million ................. 114,097 117 490,343 0.02 24,806 0.4720 Food and kindred products ................................ 500 employees ....... 5,583 143 7,070,622 0.00 288,666 0.0521 Tobacco products .............................................. 500 employees ....... 27 434 419,423,746 0.00 98,271,892 0.0022 Textile mill products ........................................... 500 employees ....... 1,306 243 4,485,467 0.00 236,814 0.1023 Apparel and other textile products .................... 500 employees ....... 4,227 49 1,717,339 0.00 84,857 0.0624 Lumber and wood products ............................... 500 employees ....... 13,854 96 1,520,435 0.00 80,494 0.1225 Furniture and fixtures ......................................... 500 employees ....... 5,860 135 2,063,881 0.00 101,980 0.1326 Paper and allied products .................................. 500 employees ....... 1,082 364 7,356,895 0.00 389,269 0.0927 Printing and publishing ...................................... 500 employees ....... 4,612 63 1,349,101 0.00 82,533 0.0828 Chemicals and allied products .......................... 500 employees ....... 3,794 388 7,758,606 0.00 573,110 0.0729 Petroleum and coal products ............................. 500 employees ....... 373 505 11,906,004 0.00 523,143 0.1030 Rubber and miscellaneous plastics products .... 500 employees ....... 3,926 192 4,132,970 0.00 252,124 0.0831 Leather and leather products ............................ 500 employees ....... 224 246 2,312,572 0.00 106,106 0.2332 Stone, clay, and glass products ........................ 500 employees ....... 5,529 209 2,337,003 0.00 101,728 0.2133 Primary metal industries .................................... 500 employees ....... 2,260 530 6,447,895 0.00 359,703 0.1534 Fabricated metal products ................................. 500 employees ....... 12,435 167 2,782,599 0.00 138,568 0.1235 Industrial machinery and equipment ................. 500 employees ....... 18,625 152 2,001,196 0.00 118,786 0.1336 Electronic and other electronic equipment ........ 500 employees ....... 4,356 237 3,836,835 0.00 184,646 0.1337 Transportation equipment .................................. 500 employees ....... 5,999 281 3,362,262 0.00 120,155 0.2338 Instruments and related products ...................... 500 employees ....... 3,266 163 3,239,263 0.00 211,242 0.0839 Miscellaneous manufacturing industries ............ 500 employees ....... 5,149 102 1,539,311 0.00 95,981 0.1140 Railroad transportation ...................................... 1500 employees ..... NA NA NA NA NA NA41 Local and interurban passenger transit ............. $5 million ................. 2,582 106 417,934 0.01 17,701 0.6042 Trucking and warehousing ................................ $18.5 million ............ 15,626 79 670,885 0.01 29,993 0.2644 Water transportation .......................................... 500 employees ....... 187 243 1,781,166 0.01 90,917 0.2745 Transportation by air .......................................... 1500 employees ..... 157 449 2,031,762 0.00 70,300 0.6446 Pipelines, except natural gas ............................ 1500 employees ..... 11 888 15,403,556 0.00 5,274,551 0.0247 Transportation services ..................................... $5 million ................. 879 55 377,507 0.02 15,544 0.3548 Communication .................................................. 1500 employees ..... 1,279 172 2,132,980 0.01 335,309 0.0549 Electric, gas, and sanitary services ................... $5 million ................. 3,809 65 883,319 0.01 72,099 0.0950 Wholesale trade—durable goods ...................... 100 employees ....... 52,553 43 1,828,263 0.00 73,131 0.0651 Wholesale trade—nondurable goods ................ 100 employees ....... 30,785 44 2,682,104 0.00 85,196 0.0552 Building materials and garden supplies ............ $5 million ................. 13,619 19 712,058 0.01 24,294 0.0853 General merchandise stores ............................. $5 million ................. 482 14 398,828 0.01 16,892 0.0854 Food stores ........................................................ $5 million ................. 6,419 140 763,042 0.00 20,647 0.6855 Automotive dealers and service stations ........... $5 million ................. 38,985 26 774,574 0.01 18,225 0.1456 Apparel and accessory stores ........................... $5 million ................. 289 41 1,346,240 0.00 85,526 0.0557 Furniture and homefurnishings stores ............... $5 milion ................. 438 71 1,685,231 0.00 87,235 0.0858 Eating and drinking places ................................ $5 million ................. 16,852 24 374,691 0.00 17,633 0.1459 Miscellaneous retail ........................................... $5 million ................. 12,619 18 406,958 0.01 22,502 0.0860 Depository institutions ........................................ $5 million ................. 788 123 1,060,910 0.00 NA NA61 Nondepository institutions .................................. $5 million ................. 840 25 728,626 0.00 106,401 0.0262 Security and commodity brokers ....................... $5 million ................. 921 33 631,139 0.01 55,488 0.0663 Insurance carriers .............................................. $5 million ................. 365 92 740,731 0.01 NA NA64 Insurance agents, brokers, and service ............ $5 million ................. 5,583 54 335,823 0.01 NA NA65 Real estate ......................................................... $5 million ................. 10,714 56 533,940 0.01 48,369 0.1267 Holding and other investment offices ................ $5 million ................. 1,036 36 889,373 0.00 95,534 0.0470 Hotels and other lodging places ........................ $5 million ................. 2,163 41 472,311 0.00 32,784 0.1372 Personal services .............................................. $5 million ................. 9,786 80 190,546 0.02 15,019 0.5373 Business services .............................................. $5 million ................. 14,343 160 517,986 0.01 37,783 0.4275 Auto repair, services, and parking ..................... $5 million ................. 43,985 47 342,341 0.01 21,749 0.2276 Miscellaneous repair services ........................... $5 million ................. 2,631 34 340,605 0.01 22,039 0.1578 Motion pictures .................................................. $5 million ................. 1,494 29 350,142 0.01 24,304 0.1279 Amusement and recreation services ................. $5 million ................. 4,052 46 469,977 0.00 23,222 0.2080 Health services .................................................. $5 million ................. 39,536 82 521,074 0.01 31,877 0.2681 Legal services .................................................... $5 million ................. 7,288 41 314,988 0.01 48,175 0.0982 Educational services .......................................... $5 million ................. 1,739 99 649,462 0.01 35,911 0.2883 Social services ................................................... $5 million ................. 5,194 43 354,060 0.01 16,245 0.2684 Museums, botanical, zoological gardens .......... $5 million ................. 158 80 492,341 0.01 35,333 0.2386 Membership organizations ................................. $5 million ................. 11,589 55 296,761 0.01 13,965 0.3987 Engineering and management services ............ $5 million ................. 11,383 62 457,931 0.01 34,480 0.1889 Services, n.e.c ................................................... $5 million ................. 679 58 423,854 0.01 36,402 0.16

1 As defined by the Small Business Administration, 61 FR 3289.2 Annual receipts.


Source: Department of Labor, Occupational Safety and Health Administration, Office of Regulatory Analysis.

Unfunded Mandates AnalysisThe final Respiratory Protection

standard has been reviewed by OSHA inaccordance with the UnfundedMandates Reform Act of 1995 (UMRA)(2 USC 1501 et seq.) and ExecutiveOrder 12875. As discussed in Chapter V,OSHA estimates that compliance withthe revised Respiratory Protectionstandard will require expenditures ofmore than $100 million each year byemployers in the private sector.Therefore, the Respiratory Protectionfinal rule establishes a Federal privatesector mandate and is a significantregulatory action within the meaning ofSection 202 of UMRA (2 U.S.C. 1532).OSHA has included this statement toaddress the anticipated effects of thefinal rule pursuant to Section 202.

OSHA standards do not apply to stateand local governments except in statesthat have voluntarily elected to adopt anOSHA State plan and have then adoptedthe specific standard in question or onethat has been deemed by OSHA to beequally effective. Consequently, theRespiratory Protection standard doesnot impose a ‘‘federal intergovernmentalmandate’’ as defined by Section 421(5)of UMRA (2 USC 658 (5)). The revisedRespiratory Protection standardtherefore does not impose an unfundedmandate on state and localgovernments.

Further, OSHA has found that thecosts incurred by state and localgovernments in those states that chooseto adopt the standard will be smallcompared to corresponding state andlocal government expenditures. If State-plan states adopt the standard, thegreatest impact in some states would beon public fire departments. Bureau ofthe Census data on the amount ofrevenue dedicated to fire protection bylocal governments indicate that $14.4billion was spent on this service in1992, the latest year for which such dataare available [Government Finances].NFPA data indicate that 75.3 percent ofthe U.S. population is served by firedepartments that employ at least somecareer firemen [NFPA, p. 15]. Thismeans that approximately 37.7 percentof the population (approximately half ofall state and local governmentemployees work in State-plan states) isserved by at least partly career firedepartments in State-plan states.Assuming the expenditures for fireprotection are spread fairly evenlyacross the population, approximately$5.3 billion is spent on fire protectionannually by affected fire departments.As indicated in the cost analysis (see

Table VI–2), the total annual cost of thestandard for public fire departments inState-plan states is approximately $3.5million, which means that the costs ofcompliance constitute less than 0.1percent of the revenue devoted by thesestates to fire protection. Costs of thismagnitude are clearly an insignificantportion of the total fire protectionbudget.

The remainder of this sectionsummarizes OSHA’s findings, asrequired by Section 202 of UMRA (2USC 1532):

This standard is issued under Section6(b) of the OSH Act.

This standard has annualized costsestimated at $111 million, primarily inthe private sector, and is estimated tosave hundreds of lives per year fromcancer and cardiovascular disease.Compliance will also prevent thousandsof illnesses annually that would havebeen caused by acute and chronicoverexposures. The standard willimpose no more than minimal costs onstate, local or tribal governments,substantially less than $100 million.OSHA pays 50 percent of State plancosts, although the Agency does notprovide funding for state, local or tribalgovernments to comply with its rules asemployers.

OSHA does not anticipate anydisproportionate budgetary effects uponany particular region of the nation orparticular state, local, or tribalgovernments, or urban or rural or othertypes of communities. The principalcosts of this standard are to controlworker exposures associated withprogrammatic provisions such as annualfit testing and training, activities that areengaged in by thousands ofestablishments in hundreds of SIC codesthat are widely distributed throughoutthe country. Chapters III and V haveprovided detailed analyses of the costsand impacts of the standard onparticular segments of the private sector.OSHA has analyzed the economicimpacts of the standard on theindustries affected and found thatcompliance costs are no more than 0.1percent of sales for establishments inany industry, and consequently that noplant closures or job losses areanticipated in the affected industries. Asa result, impacts on the nationaleconomy would be too small to bemeasurable by economic models.

Pursuant to Section 205 of the UMRA(2 USC 1535), after having considered avariety of alternatives outlined in thePreamble and in the RegulatoryFlexibility Analysis, the Agency has

concluded that the final rule is the mostcost-effective alternative forimplementation of OSHA’s statutoryobjective of reducing significant risk tothe extent feasible.

Environmental Impact AnalysisThe final Respiratory Protection

standard has been reviewed inaccordance with the requirements of theNational Environmental Policy Act(NEPA) of 1969 (42 U.S.C. 4321 et seq.),the regulations of the Council ofEnvironmental Quality (CEQ) (40 CFRpart 1500), and DOL NEPA procedures(29 CFR part 11). As a result of thisreview, OSHA has concluded that therule will have no significantenvironmental impact.

ReferencesBureau of the Census, Government

Finances, Series GF, No. 5, annual, asreported in the Statistical Abstract of theUnited States, 1995. GPO, 1995.

VII. Summary and ExplanationThis section of the preamble

summarizes and explains the provisionsof the final respiratory protectionstandard. It describes changes made tothe rule since the proposal was issued,discusses the comments received by theAgency on the proposal, and presentsOSHA’s rationale for making thesechanges. The record evidencesupporting each of the requirements ofthe final rule is also described in detailin this section.

This final rule clarifies, updates, andstrengthens OSHA’s previousrespiratory protection standard, whichwas adopted by the Agency in 1971 andhas remained essentially unchangedsince that time. This rulemaking is thusthe first major revision to OSHA’srespiratory protection standard in morethan 25 years. As discussed inconnection with several of theindividual paragraphs of the revisedstandard, not all of the provisions of thestandard have been revised; in somecases, OSHA found, and the recordsupported, leaving individualprovisions unchanged.

The final respiratory protectionstandard applies to respirator use ingeneral industry, construction,shipyards, marine terminals, andlongshoring operations. When usedproperly, respirators can help to protectemployees from the acute and chroniceffects of exposure to hazardousairborne contaminants, whether in theform of particulates, vapors, or gases.Generally, OSHA requires respirators tobe used to protect employee health in


situations where engineering controlsand work practices are not feasible,where such controls have not yet beeninstituted, in emergencies, or wheresuch controls are not sufficient, bythemselves, to protect the health ofemployees.

As noted above, this final standardapplies to respirator use in generalindustry, construction, shipyards,marine terminals, and longshoringoperations. In the 1994 proposal, OSHAproposed to cover general industry,shipyards and construction. Thelongshoring and marine terminals finalrule (48 FR 30908) already made thisstandard applicable to those industriesas well. To provide clarity, the finalrespiratory standard explicitly containsa note setting forth the scope of therespirator standard.

The preamble to the proposed ruleasked for comments about theappropriateness of applying the finalrule to construction and maritimeworkplaces. In the case of theconstruction industry, OSHAspecifically provided the AdvisoryCommittee for Construction Safety andHealth (ACCSH) with a copy of theproposal for review and comment, andACCSH recommended that the revisedstandard apply to construction industryworkplaces. OSHA’s responses to thesecomments are discussed above in theintroduction to this preamble.

In response to the question raisedabout the applicability of the standardto the construction and shipyardindustries, OSHA received severalcomments from participants concernedabout the rule’s impact on theconstruction industry (Exs. 54–102, 54–231, 54–288). These commenters notedthat the costs of the standard forconstruction employers may be higherthan for their counterparts in generalindustry because of the higher turnover,decentralization of workplaces, andmulti-employer work arrangementstypical of construction sites. However,as reported in the Final EconomicAnalysis (Ex. 196), OSHA hasdetermined that the final rule is bothtechnologically and economicallyfeasible for employers in theconstruction industry. There is noquestion that many workers in thisindustry need respiratory protection toprevent material impairment of theirhealth; in fact, some of the mosthazardous exposures occur in thisindustry. For example, workers engagedin the abrasive blasting of bridges areoften exposed to high concentrations ofsilica and other hazardous substances(contained in the abrasive blastingmedia), as well as to lead, chromates,and other toxic materials (contained in

the paints, coatings, or preservativescovering the substrate). Welders,demolition workers, tunnel workers,and painters are other examples ofconstruction trades that often involveoverexposure to toxic substances andrequire respirators for control. In fact,respirators may be even more necessaryin construction than in general industrybecause the transient and constantlychanging nature of many constructionworksites makes the use of engineeringcontrols more difficult in theseenvironments. Finally, OSHA’s previousrespiratory protection standard hasapplied to the construction industrysince 1971 (it is codified at 29 CFR1926.103); removing this protection forconstruction workers would thusdecrease existing safety and healthprotections despite the significant riskconfronting construction workers inmany situations. Decreasing feasibleworker protections in the face ofsignificant risk of material impairmentof health would clearly be contrary tothe Agency’s mandate.

OSHA received no comments on theapplicability of the final rule toshipyard employment. Likeconstruction workers, shipyard workershave been covered by the Agency’sprevious standard since 1971. Inaddition, employees in shipyardsengage in many of the same highlyhazardous operations as constructionworkers, including abrasive blasting,welding, painting, and drilling. TheFinal Economic Analysis (Ex. 196) hasdetermined that it is bothtechnologically and economicallyfeasible for employers in shipyardoperations to achieve compliance withthe final rule.

OSHA has recently issued a revisedfinal rule for the Longshoring(shipboard) portion of marine cargo-handling operations, along withrevisions to the Agency’s MarineTerminals (dockside) marine cargo-handling standard. The scope andapplication sections of both finalmaritime rules specifically incorporateOSHA’s respiratory protection standard(29 CFR 1910.134) by reference. Thus,consistent with the proposal, this finalrespiratory protection standard willapply to workplaces in general industryand in the construction, shipyards,longshoring, and marine terminalsindustries.

At the public hearing, theBrotherhood of Maintenance of WayEmployees (BMWE) submittedtestimony on the issue of OSHA’srespiratory protection standard’scoverage of railroad construction andmaintenance employees (Ex. 122). TheBMWE stated:

* * * the BMWE respectfully requests that* * * formal recognition of the applicabilityof OSHA 1910.134 for railroad employees bepublished in the Federal Register to removeany lingering questions regarding theapplicability of OSHA’s respiratoryprotection standards to working conditionswhich, although located within the railroadindustry, are in fact similar to those of anyindustrial workplace.

In response to this comment, OSHAnotes that both the prior respiratoryprotection standard and the finalrevised standard being published willapply to railway workers unless theFederal Railroad Administration (FRA)exercises statutory authority to issue aseparate respirator standard for thoseworkers. To date, the FRA has notissued a respiratory protection standardapplicable to railway workers. Unlessand until it does, this standard willapply to those workers.

This Summary and Explanationsection follows the order of the finalrule. The abbreviation ‘‘Ex.’’ denotesexhibits in the docket for thisrulemaking, Docket H–049. Theabbreviation ‘‘Tr.’’ denotes thetranscripts of the hearings conducted inconnection with this rulemaking.

Paragraph (a)—Permissible practice

Paragraphs (a)(1) and (a)(2) of the finalrule are essentially unchanged from thecorresponding paragraphs of the priorrule and the proposed rule. Indeed, inthe proposal OSHA explained that thisrulemaking was not intended to addressthe substantive portion of paragraph(a)(12). The only changes proposed byOSHA to the regulatory language ofparagraph (a) were non-substantive: (1)In the proposal, the Agency titled thisparagraph ‘‘Scope and Application’’rather than ‘‘Permissible Practice,’’which had been the title of thisparagraph since 1971; and (2) a cross-reference to paragraph (b) in the priorstandard was proposed to be changed toparagraph (c), because a new paragraph(b), ‘‘Definitions,’’ was proposed to beadded to the final rule. In the final rule,OSHA has determined that the originaltitle of paragraph (a), ‘‘PermissiblePractice,’’ better describes paragraph (a),and thus this continues to be the title ofthis paragraph. The proposed cross-reference to paragraph (c) is retained inthe final rule.

Paragraph (a)(1) requires the use ofappropriate respiratory protection when‘‘effective engineering controls are notfeasible, or while they are beinginstituted.’’ This paragraph alsostipulates that the prevention ofatmospheric contamination caused by‘‘harmful dusts, fogs, fumes, mists,gases, smokes, sprays, or vapors’’ shall


be accomplished, to the extent feasible,by the use of engineering controlmeasures.

As stated in the preamble of theproposed rule (59 FR 58895), OSHA didnot in this rulemaking open the recordon the issue of the hierarchy ofindustrial hygiene controls; thehierarchy language is merely broughtforward, verbatim, from this paragraphof the prior rule. Paragraph (a)(1), whichwas adopted by OSHA in 1971 from the1969 American National StandardsInstitute (ANSI) standard, Z88.2–1969,established that a hierarchy of controlsis to be used to protect employees fromhazardous airborne contaminants.According to this hierarchy, engineeringcontrols are the preferred method ofcompliance for protecting employeesfrom airborne contaminants and are tobe implemented first, before respiratoryprotection is used. According toparagraph (a)(1), respirators arepermitted to be used only whereengineering controls are not feasible orduring an interim period while suchcontrols are being implemented.

Paragraph (a)(2) requires employers toprovide employees with respirators‘‘when such equipment is necessary toprotect the health of the employee.’’ Inaddition, this paragraph specifies thatthe employer must provide employeeswith respirators that are ‘‘applicable andsuitable’’ for the purpose intended, i.e.,for the protection of employee health.This paragraph thus clearly recognizesthat, when properly selected, used, andmaintained, respiratory protection canplay an essential role in preventingadverse effects on the health ofemployees exposed to hazardousairborne contaminants.

By leaving paragraphs (a)(1) and (a)(2)of the final rule unchanged from thecorresponding paragraphs of therespiratory protection standard that hasbeen in effect since 1971, OSHAaccomplishes several objectives. First, itcontinues the protection that employeeshave relied on throughout OSHA’shistory. Second, it retains the languagethat employers are familiar with andthus will not require them to becomefamiliar with new regulatory language.Third, leaving the regulatory text ofparagraphs (a)(1) and (a)(2) unchangedallows OSHA and the affected public tocontinue to rely on OSHAinterpretations, decisions, and case lawthat have developed over the years.

As noted above, this standard is arespiratory protection standard. OSHAhas enforced this standard whenemployers fail to provide respirators,when the respirators that are providedare inappropriate for the form of thecontaminant or for the atmospheric

concentration of the contaminant, whenthey are inappropriately used, and whenthey are improperly maintained.

Although OSHA clearly stated in thepreamble to the proposal that thehierarchy of controls was not an issuein this rulemaking, the Agency didreceive comment on this provision. Forexample, one commenter stated that, inits opinion, OSHA has ‘‘a legalobligation to provide interested partieswith an opportunity to comment on themethods of compliance provisions’’ (Ex.54–307). In the opinion of thiscommenter, the American Iron and SteelInstitute (AISI), ‘‘Section 6(b)(2) of theOSH Act requires that OSHA provideinterested persons an opportunity tosubmit written data and comments on aproposed rule in total’’ [emphasisadded].

The unchanged language of paragraph(a)(1) was included in the proposed ruleonly to enable interested parties to viewthe rule as it would ultimately appear inthe Code of Federal Regulations in itsentirety. Since OSHA neither proposednor adopted modifications to paragraph(a)(1), the Agency believes that it is notlegally required to reconsider this issueat this time. OSHA has the authority toidentify which regulatory requirementsit is proposing to revise and whichissues are to receive regulatory priority.Limiting this rulemaking to issuesconcerning respirator programs isappropriate because such programs arethe exclusive focus of this rulemakingand to collect comments and data onadditional issues would divert resourcesfrom the task at hand.

The preference for engineeringcontrols has been reaffirmed in eachsubstance-specific health standardOSHA has published, most recently inthe Methylene Chloride standard (29CFR 1910.1052). OSHA does not believethat it is necessary or appropriate, in arulemaking dealing with respiratoryprotection, to reconsider its long-established policy with regard to thehierarchy of controls.

A number of commenters raisedanother issue in connection withparagraph (a)(1), and that is whetherbiological hazards, such as the hazardposed by exposure to Mycobacteriumtuberculosis, the infectious agent thatcauses tuberculosis (TB), are covered bythis paragraph (Exs. 54–213, 54–239,54–249). In response, OSHA emphasizesthat this respiratory protection standarddoes apply to biological hazards (seeMahone Grain Corp., 10 OSHRC 1275,1981). However, specifically with regardto the use of respirators to protectemployees from the risk of occupationalexposure to M. tuberculosis, OSHAstated at the public hearing on this

respiratory protection standard (Tr. 16–17), that the Agency’s tuberculosisstandard, which has just been proposed(62 FR 54160) would contain specificrequirements covering all aspects ofrespirator use in environments whereoccupational transmission oftuberculosis is possible. As explained inthe preamble to that standard, OSHA iscommitted to ensuring consistencybetween the respirator requirements inthe two standards.

As stated at the hearing, ‘‘until thefinal tuberculosis standard ispromulgated, we will continue toenforce respirator usage for TB underthe current, unrevised respiratorstandard, 1910.134.’’ (Tr. 18). There waslittle comment on this issue during therulemaking. The entire previousrespiratory protection standard is beingredesignated as 29 CFR 1910.139. It willbe published in the next edition of theCode of Federal Regulations under thatdesignation. OSHA’s enforcementpolicy concerning required respiratoruse for TB is set out in OSHA’sCompliance Directive, ‘‘EnforcementProcedures and Scheduling forOccupational Exposure to Tuberculosis’’(OSHA Instruction CPL 2.106). Theseenforcement procedures are based, inpart, on the Centers for Disease Controland Prevention’s (CDC) ‘‘Guidelines forPreventing the Transmission ofMycobacterium Tuberculosis in Health-Care Settings, 1994.’’ Like the CDCrecommendations, OSHA’s directiveclarifies that respiratory protection foremployees exposed to TB is requiredwhen: (1) Workers enter rooms housingindividuals with suspected orconfirmed infectious TB; (2) workers arepresent during the performance of high-hazard procedures on individuals whohave suspected or confirmed infectiousTB; and (3) emergency medical responsepersonnel or others transport, inenclosed vehicles, an individual withsuspected or confirmed infectious TB.Under the directive, OSHA also enforcesthe performance criteria recommendedby CDC for selecting a respiratorsuitable for use against TB. OSHA’sdirective further specifies that whererespirator use is required against TB, theprogram elements of OSHA’s respiratoryprotection standard apply. A copy ofOSHA’s Compliance Directive can beobtained from OSHA’s Office ofPublications (Telephone Number, 202–219–4667). Copies of the CDCGuidelines can be obtained by callingCDC (Telephone Number, 1–800–342–2437).

As noted above, paragraph (a)(2) ofthe final rule is identical both to thecorresponding paragraph of therespiratory protection standard in place


since 1971 and to proposed paragraph(a)(2). It specifies that respirators mustbe provided by the employer ‘‘whensuch equipment is necessary to protectthe health of the employee.’’ OSHAconsiders respirators to be necessary toprotect the health of the employeewhenever feasible engineering and workpractice controls are not available, arenot sufficient to protect employeehealth, have not yet been instituted, inemergencies, and where the health of anemployee is at risk (e.g., wheneveremployee exposure exceeds an OSHApermissible exposure limit (PEL)).

A violation of paragraph (a)(2) couldexist, for example, if it can be shownthat exposure to an airbornecontaminant could result in illness orinjury to the employee’s health and thatthis could be prevented by theappropriate selection and use of arespirator. An OSHA ReviewCommission case illustrates such asituation: an employer was held to haveviolated paragraph (a)(2) because hisemployees either did not use respiratorswhen working in an atmospherecontaminated with grain dust or usedrespirators that were ‘‘so caked withdust that employees could not breathethrough them’’ and contracted apotentially fatal disease caused by theinhalation of grain dust contaminatedwith Histoplasma capsulatum spores(Mahone Grain Corporation, 10 OSHRC1275, 1981). Paragraph (a)(2) was citedin this case even though OSHA has nospecific PEL for grain dust or for H.capsulatum spores.

In the past 5 years, OSHA has issued99 citations for violations of paragraph(a)(2) in conjunction with a citation ofthe General Duty Clause (i.e., Sec.5(a)(1) of the Act). These citationsconcerned various situations involvingthe failure of the employer: (1) Tocontrol exposures in emergencies; (2) tocontrol exposure to unknownconcentrations of a toxic substance; (3)to control exposure to a contaminantthat was clearly a recognized hazardeven though no OSHA PEL existed; (4)to provide and require the use of arespirator for a confined space entry; or(5) to ensure the proper use of arespirator in a situation involving theimproper storage of a chemical(s).OSHA will continue to view thesesituations as citable under this standardbecause they involve failure toimplement the appropriate exposurecontrol necessary to protect the healthof the employee from adverse effects.

As proposed, paragraph (a)(3) ofOSHA’s prior standard does not appearin the final rule. This paragraph, whichwas adopted by OSHA in 1971 from theANSI Z88.2–1969 standard, stated that

employees must use the respiratoryprotection provided in accordance withinstructions and training they havereceived.

Several commenters (Exs. 54–79, 54–181, 54–226, 54–234, 54–295, 54–307,54–334) urged OSHA to retain thisparagraph in the final rule. According tothese commenters, this paragraph isnecessary to ensure that employees takeresponsibility for their actions and thatemployees are actively involved in therespirator program and conform toprogram procedures. OSHA agrees thatactive employee involvement in therespirator program is essential toprogram effectiveness but does notbelieve that this principle should bestated in the standard, for a number ofreasons. First, the OSH Act itself, at Sec.5(b), states that ‘‘Each employee shallcomply with occupational safety andhealth standards and all rules,regulations, and orders issued pursuantto the OSH Act which are applicable tohis own actions and conduct.’’ Inaddition, the courts have repeatedlyheld that employers are responsibleunder Section 5(a)(2) of the Act (29U.S.C. 654(a)(2)) for ensuring workerprotection (see, e.g., Brock v. City OilWell Service Co., 795 F.2d 507, 511 (5thCir. 1986)). In this case, the court held,‘‘it is the employer’s responsibility toensure that the employees are protected.It may accomplish this objectivethrough others if it chooses, but the dutyto provide the protection remains theemployer’s.’’ Accordingly, the final ruledoes not contain this paragraph.

An issue raised by OSHA inconnection with paragraph (a) of theproposal, the use of respirators byemployees when such use is required byan individual employer or is chosenvoluntarily by employees but notmandated by OSHA in this final rule, isaddressed below in connection withparagraph (c) of this Summary andExplanation.

Paragraph (b)—DefinitionsThe final standard includes

definitions of important terms used inthe regulatory text of the final rule. Theprevious and proposed respiratoryprotection standards contained nodefinitions; however, OSHA is adding anumber of definitions to the final rulebecause the Agency believes thatemployers and employees will benefitfrom this additional information. This isconsistent with the Agency’s desire toclarify its respiratory protectionrequirements, including those that arenot being substantively changed in thisrulemaking.

A number of the definitions relate tospecific types of respiratory protection

devices or to components or designcharacteristics of those devices. Forexample, the terms ‘‘air-purifyingrespirator,’’ ‘‘filter or air-purifyingelement,’’ and ‘‘positive pressurerespirator’’ are defined in the final rule.These definitions, which are derivedfrom generally recognized sources suchas the current ANSI Z88.2–1992respiratory protection standard, theNIOSH requirements for particulaterespirators in 42 CFR part 84, and the1987 NIOSH Respirator Decision Logic(Ex. 38–20), have been revised forclarity, consistency with complianceinterpretations of the Agency’srespiratory protection standard, and torespond to comments received duringthe rulemaking.

A number of commenters (Exs. 54–208, 54–218, 54–219, 54–410, 54–424)suggested that OSHA adopt several ofthe definitions in the ANSI Z88.2–1992respiratory protection standard. Theregulated community is already familiarwith the ANSI definitions of theseterms, and OSHA agrees that thepotential for confusion will be reducedif terms mean the same thing in both theOSHA and ANSI standards. Therefore,the ANSI definitions of ‘‘airlinerespirator (supplied-air respirator orairline respirator),’’ ‘‘canister orcartridge,’’ ‘‘demand respirator,’’ ‘‘end-of-service-life indicator,’’ ‘‘escape-onlyrespirator,’’ ‘‘filter,’’ ‘‘fit check (user sealcheck),’’ ‘‘fit test,’’ ‘‘helmet,’’ ‘‘hood,’’‘‘loose-fitting facepiece,’’ ‘‘negativepressure respirator,’’ ‘‘pressure demandrespirator,’’ ‘‘powered air-purifyingrespirator (PAPR),’’ ‘‘respiratory inletcovering,’’ ‘‘self contained breathingapparatus (SCBA),’’ ‘‘service life,’’ and‘‘tight-fitting facepiece’’ have all beenadded to the final standard, with someminor word changes to improve clarityand to recognize the mandatory natureof OSHA standards. In other cases,OSHA has substituted an ANSIdefinition for one the Agency originallyproposed.

Several commenters urged OSHA toadd other definitions to those in theproposal (Exs. 54–208, 54–218, 54–219,54–222, 54–251 54–267, 54–283, 54–289, 54–363, 54–410, 54–437, 54–455).OSHA did not add some of thesuggested definitions, such as one for‘‘health screening,’’ because the term isno longer used in the standard. Otherterms, such as ‘‘medical evaluation,’’ aredefined where they appear in theregulatory text.

The following discussion addresseschanges made since the proposedstandard.

Adequate warning properties. Theproposed definition of ‘‘adequatewarning properties’’ has not been


retained in the final standard becausethe term is no longer used in theregulatory text. OSHA deleted the termafter concluding that the two majorwarning properties, odor and irritation,are unreliable or inappropriate to use asindicators of sorbent exhaustion. Thisissue is discussed further in thisSummary and Explanation inconnection with paragraph (d).

Air-purifying respirator. The finalstandard defines the term ‘‘air-purifyingrespirator’’ as ‘‘a respirator with an air-purifying filter, cartridge, or canisterthat removes specific air contaminantsby passing ambient air through the air-purifying element.’’ Marc Evans ofBaxter Diagnostics, Inc. (Ex. 54–38)stated that the proposed definition, ‘‘arespirator which is designed to removeair contaminants [i.e., dust, fumes,mists, gases, vapors, or aerosols] fromthe ambient air or air surrounding therespirator,’’ was inaccurate since filterelements can only remove aircontaminants when air passes throughthe filters; he stated that the ANSIdefinition was more accurate in thisregard.

Another commenter wanted to addthe term ‘‘biologicals’’ to the list of aircontaminants removed by air-purifyingrespirators (Ex. 54–249). In response,the definition has been revised to statemore clearly that an air-purifyingrespirator removes specificcontaminants from the ambient air bydrawing air through appropriate filters,cartridges, or canisters. Deleting theproposed definition’s examples of aircontaminants makes clear that no typeof air contaminant, including biologicalagents, is excluded from the definition.Also, the term ‘‘filter’’ has been changedto ‘‘filter or air-purifying element,’’which is also defined in the standard,and includes the broad range of filters,cartridges, canisters and other air-purifying elements used withrespirators.

Assigned protection factor. Thedefinition of ‘‘assigned protectionfactor’’ has been reserved as part ofOSHA’s decision to address the entireAssigned Protection Factor (APF) issuein a subsequent phase of thisrulemaking. OSHA proposed toreference the NIOSH assignedprotection factors from the 1987 NIOSHRespirator Decision Logic in therespiratory protection standard and thento adopt new APF values issued byNIOSH after that Agency had conductedrulemaking on APFs. In the course ofthis rulemaking, OSHA has concludedthat it should instead develop its ownset of assigned protection factors basedon a thorough review and analysis of allrelevant evidence. Both the NIOSH and

the ANSI APFs, as well as all relevantdata and information, will beconsidered by OSHA at that time.

Atmosphere-supplying respirator.This term means ‘‘a respirator thatsupplies the respirator user withbreathing air from a source independentof the ambient atmosphere, andincludes supplied-air respirators (SARs)and self-contained breathing apparatus(SCBA) units.’’ As it has done in manyof the definitions in this section, OSHAhas substituted the term ‘‘breathing air’’for a number of synonymous, butconfusingly diverse, terms used in theproposal and in the ANSI Z88.2–1992standard. The minor changes from theproposed definition have been madesolely to enhance clarity.

Canister or cartridge. The finalstandard adopts the ANSI Z88.2–1992standard’s definition: ‘‘a container witha filter, sorbent, or catalyst, orcombination of these items, whichremoves specific contaminants from theair passed through the container.’’Several commenters suggested that thisdefinition be added to the final rule(Exs. 54–208, 54–218, 54–219, 54–410,54–424).

Demand respirator is defined as ‘‘anatmosphere-supplying respirator thatadmits breathing air to the facepieceonly when a negative pressure is createdinside the facepiece by inhalation.’’ Thisterm was not defined in the proposalbut is defined by ANSI, and severalcommenters (Exs. 54–208, 54–218, 54–219, 54–410, 54–424) urged that it beincluded in the final rule. As in otherdefinitions, the phrase ‘‘breathing air’’has been substituted for ‘‘respirable gas’’for clarity.

The proposal’s definition of‘‘demand’’ has been deleted from thefinal standard because the addition of adefinition for ‘‘demand respirator’’makes its inclusion unnecessary. (Seethe definition of pressure demandrespirator below for the distinctionbetween the two types of respirator.)

Dust mask. See the definition for‘‘filtering facepiece’’ below.

Emergency situation. In the final rule,OSHA is adding this term to paragraph(b) to clarify its use in the regulatorytext. ‘‘Emergency situation’’ is definedas ‘‘any occurrence such as, but notlimited to, equipment failure, rupture ofcontainers, or failure of controlequipment that may or does result in anuncontrolled substantial release of anairborne contaminant.’’ Under thisdefinition, OSHA intends that apotential release, and not just an actualrelease, be considered an emergencysituation requiring appropriaterespiratory protection. This definition isthe same or similar to those used to

define emergency situations in otherOSHA health standards (e.g., 1910.1051,Butadiene; 1910.1028, Benzene;1910.1048, Formaldehyde).

Employee Exposure. OSHA has addedthis term to paragraph (b) of the finalrule and has defined it to mean‘‘exposure to a concentration of anairborne contaminant that would occurif the employee were not usingrespiratory protection.’’ This is the samedefinition that has been used in manyof OSHA’s substance-specific healthstandards. It is included to clarify thatemployee exposure is measured outsideany respiratory protection worn.

End-of-service-life indicator (ESLI)means ‘‘a system that warns therespirator user of the approach of theend of adequate respiratory protection,for example, that the sorbent isapproaching saturation or is no longereffective.’’ This definition was not in theproposal, but has been derived from thedefinition in the ANSI Z88.2–1992standard, as requested by severalcommenters (Exs. 54–208, 54–218, 54–219, 54–410, 54–424). OSHA hasincluded the example at the end of thedefinition to clarify the function of anESLI.

Escape-only respirator. This term wasnot defined in the proposal, but the finalstandard defines an escape-onlyrespirator as ‘‘a respirator intended to beused only for emergency exit.’’ The DowChemical Company (Ex. 54–278) andthe Chlorine Institute (Ex. 54–439)recommended adding definitions for an‘‘escape’’ respirator and an ‘‘emergency’’respirator. Partially in response to thesecomments, and to clarify OSHA’s intent,OSHA has described in paragraph (d)the narrow function of an ‘‘escape-onlyrespirator,’’ and has added a definitionfor ‘‘escape-only respirator’’ to thisparagraph (b). The definition of ‘‘escape-only respirator’’ derives from the ANSIZ88.2–1992 standard, with the phrase‘‘egress from a hazardous atmosphere’’replaced by the word ‘‘exit.’’

Filter or air-purifying element. Thefinal standard’s definition of this term is‘‘a component used in respirators toremove solid or liquid aerosols from theinspired air.’’ The parallel definition inthe proposal used ‘‘filter’’ instead of‘‘filter or air-purifying element’’ and hasbeen changed in response to comments(Exs. 54–208, 54–218, 54–219, 54–410,54–424). The phrase ‘‘or air-purifyingelement’’ has been added to clarify thatthis definition applies to all filtrationmechanisms, not only to mechanical orelectrostatic filtration of particulates.The new definition derives from thedefinition of ‘‘filter’’ in the ANSI Z88.2–1992 standard.


Filtering facepiece (dust mask). Thedefinition of ‘‘filtering facepiece’’ in thefinal rule is ‘‘a negative pressureparticulate respirator with a filter as anintegral part of the facepiece or with theentire facepiece composed of thefiltering medium.’’ This new definitionis derived from the definition of‘‘filtering facepiece’’ in the NIOSHRespirator Decision Logic (Ex. 38–20).As described in the discussion ofparagraph (c) below, employers whoallow the use of these respirators whensuch use is not required need to complywith only paragraph (c)(2) of thisstandard, which requires that theemployer provide the employee withthe information contained in AppendixD.

Fit factor. The definition of ‘‘fitfactor’’ in the final rule is a quantitativeestimate of the fit of a particularrespirator to a specific individual, andtypically estimates the ratio of theconcentration of a substance in ambientair to its concentration inside therespirator when worn. In the proposal,OSHA’s definition included the terms‘‘challenge agent’’ and ‘‘test chamber.’’Several commenters (Baxter Diagnostics,Ex. 54–38; American SubcontractorsAssociation, Ex. 54–293) stated thatusing these terms would have theunintended effect of prohibiting the useof several existing QNFT test methods,such as the TSI Portacount,TM andrecommended that OSHA rely on theANSI definition of ‘‘fit factor’’ instead.OSHA agrees with this point, and thefinal standard’s definition derivesprimarily from the ANSI Z88.2–1992standard’s definition, as commenterssuggested (Exs. 54–208, 54–218, 54–219,54–410, 54–424). The final definitionuses the word ‘‘estimate’’ instead of theANSI definition’s word ‘‘measure’’because fit factors estimate, rather thanmeasure, the fit obtained during use.The phrase ‘‘specific individual’’ hasbeen substituted for ‘‘particularindividual’’ for clarity.

Fit test. A definition of ‘‘fit test’’ hasbeen added to the final rule and isdefined as ‘‘the use of a protocol toqualitatively or quantitatively evaluatethe fit of a respirator on an individual.’’(See also QLFT and QNFT.) Thisdefinition has been added becauseOSHA is of the opinion, based oncomments to the record, that such adefinition is needed (Exs. 54–208, 54–218, 54–219, 54–410, 54–424). ANSIalso has a definition of fit test, butOSHA’s definition differs from that inthe ANSI Z88.2–1992 standard in thatthe term ‘‘challenge agent’’ has beeneliminated and replaced by the phrase‘‘protocol to quantitatively orqualitatively evaluate.’’ The use of the

term ‘‘challenge agent’’ would limit thedevelopment of future fit testtechnologies that do not involve a testagent (Exs. 54–208, 54–250, 54–330, 54–424).

Hazardous exposure level. Becausethe final standard does not use the term‘‘hazardous exposure level,’’ it is notdefined. The proposal defined suchlevels as including the PermissibleExposure Limits (PELs) contained inOSHA’s Tables Z–1, Z–2, and Z–3 of 29CFR 1910.1000; the AmericanConference of Governmental IndustrialHygienists (ACGIH) Threshold LimitValues (TLVs), as published in the latestedition of that organization’s‘‘Threshold Limit Values for ChemicalSubstances and Physical Agents,’’ forthose substances without an OSHA PEL;the NIOSH Recommended ExposureLimits (RELs) for those hazardouschemicals without either an OSHA PELor ACGIH TLV; and any exposure levelbased on available scientificinformation, including Material SafetyData Sheets, for those hazardouschemicals for which no OSHA PEL,ACGIH TLV, or NIOSH REL has yetbeen published.

The proposed rule would haverequired employers to identify the‘‘hazardous exposure level’’ applicableto each hazardous chemical in theworkplace and then to use thisinformation in selecting the appropriaterespirator to provide protection againstexposure to that chemical. The final ruletakes a different and much simplerapproach to assisting employers in theselection of appropriately protectiverespirators in those cases where OSHAhas not yet promulgated a PEL for ahazardous chemical. OSHA has takenthe approach reflected in the finalstandard because there was widespreadobjection to the proposed approach(Exs. 54–94, 54–175, 54–212, 54–226,54–232, 54–275x, 54–283, 54–293, 54–306, 54–312, 54–324, 54–334, 54–347,54–352, 54–361, 54–397, 54–443, 54–445). Some commenters (Exs. 54–91,54–165, 54–181, 54–291, 54–316, 54–347, 54–397, 54–445) interpreted theproposed approach as an attempt byOSHA to expand the number ofhazardous chemicals with OSHA-enforceable exposure limits, whileothers believed that implementing theproposed approach would requireemployers to have risk assessmentexpertise or to perform complexanalyses, and pointed out that manyemployers lacked such expertise (Exs.54–106, 54–175, 54–210). In general,rulemaking participants stated thatOSHA’s approach to this problemshould rely on the professionaljudgment of employers, based on readily

available information (Exs. 54–206, 54–210).

OSHA has decided, after a thoroughreview of the record, to follow theserecommendations, and in the final rulehas adopted an approach that requiresemployers to select appropriatelyprotective respirators on the basis ofinformed professional judgment.Accordingly, the final rule does notidentify the ACGIH TLVs or the NIOSHRELs as references that would triggerrequired respirator use. The approachtaken in the final rule providesemployers with the flexibility to rely onprofessional judgment and availabledata sources when selecting respiratorsfor protection against hazardouschemicals that have no OSHA PEL.

OSHA believes that it is prudent insuch cases for employers to select morerather than less protective respirators,i.e., to select a respirator that willreduce employee exposure to a levelbelow the concentration indicated ashazardous by the scientific literature.OSHA also believes that manyemployers will choose to rely on theACGIH TLV or NIOSH REL in thosecases where OSHA has no PEL at thepresent time. However, whateverapproach employers choose to take, therespirator selected must ‘‘be applicableand suitable for the purpose intended,’’as required by paragraph (a).

Helmet. The final standard defines ahelmet as ‘‘a rigid respiratory inletcovering that also provides headprotection against impact andpenetration.’’ This definition, whichwas not in the proposal, has been addedto the final standard at the request ofseveral commenters ( Exs. 54–208, 54–218, 54–219, 54–410, and 54–424). TheOSHA definition uses the term‘‘respiratory inlet covering’’ instead ofthe word ‘‘hood’’ used in the ANSIdefinition in order to include helmet-style powered air-purifying respirators(PAPRs).

High efficiency particulate air (HEPA)filter is defined as ‘‘a filter that is at least99.97% efficient in removingmonodisperse particles of 0.3micrometers in diameter. Theequivalent NIOSH 42 CFR 84 particulatefilters are the N100, R100, and P100filters.’’ Although NIOSH has revisedthe particulate filter descriptions underthe new 42 CFR Part 84 respiratorcertification regulation, and no longeruses the term HEPA, this definition isincluded because ‘‘HEPA filter’’ is usedin many of OSHA’s substance-specificstandards. The definition, which issimilar to that used by ANSI, lists theNIOSH 42 CFR part 84 particulate filtersthat are equivalent, in terms of


efficiency, to the HEPA filter, i.e., theN100, R100, and P100 filters.

Hood. The final standard includes thefollowing definition of ‘‘hood’’: ‘‘arespiratory inlet covering thatcompletely covers the head and neckand may also cover portions of theshoulders and torso.’’ This definitionhas been added to the final standard inresponse to commenters (Exs. 54–208,54–218, 54–219, 54–410, and 54–424).The definition derives from the ANSIZ88.2–1992 standard; the word ‘‘also’’has been added for clarity.

Immediately dangerous to life orhealth (IDLH). The final standarddefines IDLH as ‘‘an atmosphere thatposes an immediate threat to life, wouldcause irreversible adverse health effects,or would impair an individual’s abilityto escape from a dangerousatmosphere.’’ In the proposal, thedefinition of IDLH was ‘‘an atmosphericconcentration of any toxic, corrosive, orasphyxiant substance that poses animmediate threat to life or would causeirreversible or delayed adverse healtheffects or would interfere with anindividual’s ability to escape from adangerous atmosphere.’’ In the finalrule, OSHA has decided that includingall atmospheres capable of causing thelisted health effects is more consistentwith OSHA’s intent than limiting thedefinition to toxic, corrosive, andasphyxiant atmospheres and has alsodeleted the word ‘‘delayed’’ from thedefinition because including it causedconsiderable confusion amongcommenters.

Under the final standard’s definition,atmospheres where a short, one-timeexposure (i.e., an acute exposure) maycause death or irreversible adversehealth effects immediately, within a fewhours, or within a few days or weeks areconsidered IDLH atmospheres. Theseverity of the adverse effects and thecertainty that health impairment willoccur following an acute exposure aremore important considerations indefining a potential IDLH situation thanis the time course of the health effect.For example, an atmosphere containinglife-threatening or health-impairingconcentrations of fluorides, cadmiumfumes, or radioactive substances wouldbe considered IDLH even though asingle exposure might not cause deathor permanent impairment for as long asdays or even weeks after the exposure.On the other hand, many situationsinvolving atmospheres exceeding short-term or ceiling exposure limits are notIDLH atmospheres; most short-term orceiling limits are designed to reduce therisk of less serious effects, such assensory irritation. Thus, only thosesituations where the acute exposure

would threaten life, initiate anirreversible process that threatens life orhealth, or impede the ability of theworker to escape from the atmospherewould constitute IDLH conditions. Incontrast, if chronic exposure to a toxicatmosphere is required to producehealth impairment or cause death, theatmosphere is not IDLH. Thus, therelatively low atmosphericconcentrations of carcinogenicsubstances that cause work-relatedcancers are not considered IDLHatmospheres, even though the effect oflong-term exposure at suchconcentrations is death or seriousillness.

Paragraphs (d) and (g) of the finalstandard require employers whoseemployees are exposed to an IDLHatmosphere to provide them with themost protective and reliable respiratoryprotection, i.e., a full facepiece pressuredemand SCBA certified by NIOSH for aminimum of a 30-minute service life, ora combination full facepiece pressuredemand supplied-air respirator withauxiliary self-contained air supply, andto implement specific rescueprecautions and communicationprocedures. Although OSHA’s priorRespiratory Protection standard doesnot explicitly use the term ‘‘IDLH,’’ itdoes require that respirators used in‘‘immediately dangerous’’ atmosphereskeep inward leakage to a minimum andbe highly reliable (See paragraph (c) ofprior 29 CFR 1910.134, whichincorporates this language from theANSI Z88.2–1969 standard byreference).

Commenters raised a number of issuesspecifically related to the proposeddefinition of IDLH and to the IDLHconcept in general. These commentsaddressed the following points:• Whether the term IDLH should apply

to all delayed effects, some delayedeffects, or be restricted to immediateeffects;

• How OSHA’s definition of IDLHdiffers from those of otherorganizations and how it relates to thedefinition of IDLH used in otherOSHA standards;

• How the presence of an IDLH orpotential IDLH atmosphere affectsrespirator selection.

The following discussion addresseseach of these points in turn.

The proposed definition of IDLHincluded the phrase ‘‘delayed adversehealth effects.’’ OSHA has omitted thisphrase from the final standard torespond to comments received and toremove a source of confusion. Manycommenters argued that the term IDLHshould cover only immediate, severe

adverse health effects, such as thoseresulting from exposures to hydrogenfluoride or oxides of nitrogen (e.g., Exs.54–208, 54–219; 54–316), while othersfavored taking chronic, delayed effectsinto consideration when making anIDLH decision (See, e.g., Exs. 54–202and 54–437). For example, OCAWstated that ‘‘OSHA’s IDLH and acutehazard-based framework * * * does notproperly emphasize the need toconsider long-term and cumulativehealth effects.’’

Most participants, however, arguedagainst including chronic health effectsin the IDLH definition because it wouldmake the definition too broad. Theseparticipants feared that including thisterm would mean that exposurestypically associated with chroniceffects, such as cancer, would bedesignated IDLH (Exs. 54–67; 54–153;54–175; 54–208; 54–218; 54–219; 54–232; 54–266; 54–278; 54–307; 54–314;54–316; 54–326). Typical of thesecomments is one from the AmericanIron and Steel Institute: ‘‘The proposeddefinition, which includes ‘‘delayedhealth effects,’’ is so broad that it goesfar beyond the accepted IDLH concept,and would expand it beyond itsintended purpose’’ (Ex. 54–307).Arguing along the same lines, the ExxonCorporation stated that ‘‘the phrase‘delayed health effects’ could includechronic toxins like asbestos * * *’’ (Ex.54–266).

Other commenters urged OSHA tonarrow the definition of IDLH by addingthe word ‘‘acute’’ before ‘‘adverse’’ inthe phrase ‘‘delayed adverse healtheffects’’ or by making other languagechanges that would achieve the sameeffect (Exs. 54–67, 54–278, 54–326, 54–208A). For example, the AmericanIndustrial Hygiene Association (Ex. 54–208A) stated that the only atmosphericcontaminants with delayed effects thatshould be included in the definition arethose, such as the oxides of nitrogen,that cause delayed-onset severe adversehealth effects (such as pulmonaryedema). Representatives of Pennzoilsuggested that ‘‘* * * the phrase‘immediate or delayed irreversibledebilitating health effects’, be used’’ toachieve the same end (Ex. 54–287).

These commenters objected to theinclusion of ‘‘delayed health effects’’ inthe proposed definition because thelanguage suggested that effects typicallyassociated with long-term exposures,such as cancer, would be included. Thedefinition in the final standardrecognizes that the effects of concernmust be the result of an acuteoverexposure but does not specificallylimit the length of time between thatoverexposure and the resulting effect.


Where very serious health effects mayarise from a single acute exposure, evenif such effects become apparent onlyafter a relatively long latency period,e.g., hours, days, or even weeks, theatmosphere associated with the effectmust be designated IDLH. OSHA isconfident that deleting the word‘‘delayed’’ from the IDLH definition inthe final rule will reduce confusion butwill not affect the level of employeeprotection provided by the standard.

Many commenters urged OSHA toadopt an IDLH definition developed byanother organization, agency, or byOSHA itself in other standards. Somecommenters (Exs. 54–153, 54–214, 54–234, 54–251, 54–266, 54–278, 54–290,54–330, 54–361, 54–363, 54–424, 54–439) urged OSHA to adopt the ANSIZ88.2–1992 standard’s definition ofIDLH: ‘‘any atmosphere that poses animmediate hazard to life or posesimmediate irreversible debilitatingeffects on health’’ (clause 3.33). Forexample, Bell Atlantic (Ex. 54–361)suggested that the ANSI definition beused to ensure that ‘‘chronic toxins likeasbestos would not be consideredIDLH.’’ However, OSHA believes thatadopting the definition contained in thecurrent ANSI standard could reduceemployee protection because it statesthat atmospheres are IDLH only in caseswhere the adverse effects of exposureoccur immediately. An example of anatmosphere that OSHA believes must beconsidered IDLH but arguably wouldnot be so designated under the ANSIdefinition is one containing highconcentrations of cadmium fume, whichmay result in fatal collapse as long as48–72 hours after an acuteoverexposure.

The Exxon Corporation (Ex. 54–266)objected to the phrase ‘‘ability toescape’’ in OSHA’s proposed definition,and suggested that OSHA instead adoptthe ANSI definition, which does notrefer to impairment of the ability toescape. OSHA wishes to clarify that theproposed terminology, ‘‘interfere withan individual’s ability to escape’’ wasnot meant to cover a minor or evenmoderate degree of interference but toaddress interference of a kindsufficiently serious to impair theindividual’s ability to escape fromexposure to a dangerous concentrationof an air contaminant. To addressExxon’s concern, the final rule’sdefinition has been revised to read‘‘impair the individual’s ability toescape.’’ OSHA notes that it isimperative for employees to be able toescape. There are atmospheres, forexample one contaminated with asevere eye irritant, that can effectivelyincapacitate an individual in the short

term and prevent the individual fromescaping in time to avoid more serioushealth consequences. OSHA hastherefore retained in the IDLH definitionlanguage that addresses the need toprotect workers escaping fromdangerous atmospheres.

One commenter, Monsanto (Ex. 54–219), expressed concern about theconsistency of IDLH definitions indifferent OSHA standards. In response,OSHA has reviewed the definitions ofIDLH used in its standards and believesthat the final standard’s definition islargely consistent with those in the twoOSHA safety standards that use theterm: 29 CFR 1910.146, the Permit-Required Confined Space standard(‘‘Confined Spaces standard’’) and 29CFR 1910.120, the Hazardous WasteOperations and Emergency Response(HAZWOPER) standard.

Some commenters (Exs. 54–439, 54–330, 54–278) asked which IDLH valuesOSHA endorses or pointed to thelimitations of the available informationon IDLH concentrations. For example,OCAW noted that ‘‘only a handful ofIDLH limits have been determined. Inmost worker exposure, the IDLH limit isunknown. Even when [an] IDLH limitexists, workers do not have access tothis information. MSDSs rarely includeIDLH information’’ (Ex. 54–202).

The final rule does not contain aprescribed list of IDLH values or requireemployers to rely on any particular list.Some commenters (Exs. 54–278, 54–330, 54–361, 54–424, 54–439) criticizedthe IDLH values listed in the 1994NIOSH Pocket Guide to ChemicalHazards (Ex. 54–278) or recommendedthat the Emergency Response PlanningGuidelines (ERPGs) developed underthe auspices of the American IndustrialHygiene Association be used instead.OSHA is aware that published IDLHvalues are not available for manyindustrial contaminants and thatemployers must therefore rely on theirown knowledge and judgment, and thatof safety and health professionals, whendeciding that a given atmosphere hasthe potential to cause health effects ofthe kind envisioned by OSHA’s IDLHdefinition. During enforcementinspections, OSHA will continue toaccept any published IDLH value that isbased on sound scientific evidence;those published by NIOSH and theAIHA would clearly meet this test.

OSHA’s final IDLH definition doesnot separately mention ‘‘potential’’IDLH atmospheres. Many OSHAenforcement cases have involved thefailure of employers to providerespirators in situations that were notIDLH at the time workers entered thearea but became so thereafter. OSHA

intends employers to interpret therespirator selection requirements inparagraph (d)(1) proactively, i.e., whereemployers are uncertain about theadequacy of a given respirator for ahighly hazardous atmosphere, cannotidentify the atmospheric concentrationof a substance that poses a potentiallylife-threatening or health-impairing risk,or cannot maintain the concentration ofsuch a substance below life-threateningor health-impairing levels, the employermust consider the atmosphere IDLH andselect a respirator accordingly. Forexample, an employer in a chemicalplant knows that inadvertent releases orspills of highly hazardous chemicalsmay occur at the facility and selects themost protective respirators available foremployees who must enter a spill areabecause, in an emergency, there is notime to take airborne measurements todetermine whether or not theconcentration is IDLH. OSHAencourages this kind of proactiveplanning because it is protective ofemployee health.

Interior structural firefighting. Thefinal respiratory protection standarduses the OSHA definition for ‘‘interiorstructural firefighting’’ contained in 29CFR 1910.155, which applies to allsituations covered by Subpart L—FireProtection. The definition is as follows:

Interior structural firefighting means thephysical activity of fire suppression, rescueor both, inside of buildings or enclosedstructures which are involved in a firesituation beyond the incipient stage.

Loose-fitting facepiece. The finalstandard now defines this term to mean‘‘a respiratory inlet covering that isdesigned to form a partial seal with theface.’’ This definition was not in theproposal, and has been added inresponse to commenters such as theAIHA (Ex. 54–208), 3M (Ex. 54–218),Monsanto (Ex. 54–219), Martin MariettaEnergy Systems, Inc. (Ex. 54–410), andORC (Ex. 54–424), who recommendedthat OSHA adopt several of the ANSIZ88.2–1992 definitions for respiratorterms. OSHA has adopted only part ofthe ANSI definition for loose-fittingfacepiece. The phrase in the ANSIdefinition that states a loose-fittingfacepiece ‘‘does not cover the neck andshoulders, and may or may not offerhead protection against impact andpenetration’’ has not been included.This phrase from the ANSI definitionwas not adopted as part of the OSHAdefinition because adding this phrasewould not allow users to clearlydistinguish between hoods, helmets,and loose-fitting respirators. It isimportant for employers to be able todistinguish loose-fitting from tight-


fitting respirators in order to correctlyapply the fit testing requirements.

Maximum use concentration. OSHAis not defining this term at this timebecause the Agency has reserved theissue of Assigned Protection Factors,which is associated with Maximum UseConcentrations, until a subsequentphase of this rulemaking.

Negative pressure respirator (tightfitting). The final standard defines thisterm as ‘‘a respirator in which the airpressure inside the facepiece is negativeduring inhalation with respect to theambient air pressure outside therespirator.’’ The proposed definitionwas revised in response to comments(Exs. 54–208, 54–218, 54–219, 54–410,and 54–424) that recommended thatOSHA adopt the ANSI Z88.2–1992standard’s definition. In the final rule,OSHA has accepted the ANSIdefinition, with two changes: (1) Theword ‘‘facepiece’’ has replaced the term‘‘respiratory inlet covering’’ to makeclear that the facepiece is the area ofinterest with negative pressurerespirators; and (2) the phrase ‘‘outsidethe respirator’’ has been added after thephrase ‘‘ambient air pressure’’ to clarifythat negative pressure exists only whenthe outside air pressure is higher thanthe air pressure inside the negativepressure facepiece.

Oxygen-deficient atmosphere. Theproposed definition of an ‘‘oxygendeficient atmosphere’’ was ‘‘anatmosphere with an oxygen content ofless than 19.5% by volume at altitudesof 8000 feet or below.’’ OSHA isretaining the 19.5% definition of anoxygen-deficient atmosphere in the finalrule, but is removing the reference toaltitudes. The use of a 19.5% oxygenlevel is well established and has evenbeen incorporated by Congress intoother safety and health legislation (SeeFederal Mine Safety and Health Act, 20USC 863 (b), discussed in NationalMining Association v. MSHA, 116 F.3d520 (D.C. Cir. 1997.) Paragraph d(2)(iii)of the final rule requires employers toconsider all oxygen-deficientatmospheres to be IDLH and to requirethe use of pressure-demand SCBA or acombination full-facepiece pressure-demand SAR with an auxiliary self-contained air supply. However, thisparagraph also contains an exceptionthat would permit employers to use anyatmosphere-supplying respirator inoxygen-deficient atmospheres where theemployer can demonstrate that oxygenlevels cannot fall below the altitude-adjusted concentrations prescribed inTable II of paragraph (d).

The ANSI Z88.2–1992 standard,NIOSH (Ex.164), and AIHA (Ex. 2098)use an altitude-adjusted definition for

oxygen deficiency. Although there aresome small differences, theseorganizations generally define oxygendeficiency as an oxygen level of lessthan 19.5% at altitudes up to 5,000 or6,000 feet, and less than 20.9% at higherelevations. OSHA chose not to adoptthis approach to defining oxygendeficiency for several reason. First, aswas stated in the proposal (59 FR58905), OSHA’s concern is thatemployees not be exposed toenvironments in which the oxygenpartial pressure is less than 100 mm Hg;this partial pressure of oxygen isgenerally regarded as an appropriateIDLH level (Exs. 164, 208). OSHAbelieves that using an oxygenconcentration of 19.5 percent as abaseline oxygen level is appropriatebecause exposure to such an atmospheredoes not pose a serious health risk atelevations below 8,000 feet, i.e., theoxygen partial pressure in suchatmospheres will remain above 100 mmHg (Ex.164). Although OSHA realizesthat the partial pressure of oxygen maybe at or above 100 mm Hg even at somelower altitudes and lower oxygenconcentrations, these lower-altitude,lower-concentration situations aregenerally unstable and can quicklydeteriorate to life-threateningatmospheres. OSHA has accounted forthose rare situations where theemployer controls the environment tomaintain a constant altitude-adjustedoxygen level through the exception inparagraph (d)(2)(iii) of the final rule.OSHA’s definition of oxygen deficiencyis also consistent with the CompressedGas Association’s definition of Grade Dbreathing air as air containing aminimum of 19.5% oxygen. OSHA findsthat defining oxygen deficiency as anatmosphere with an oxygen contentbelow 19.5% is both protective andstraightforward, and is consistent withthe definition that has been used by theAgency in the past.

Oxygen-deficient IDLH atmosphere.The proposal originally included adefinition of oxygen-deficient IDLHatmosphere. Because the term has notbeen used in the regulatory text of thefinal rule, OSHA is deleting this termfrom paragraph (b).

Physician or other licensed healthcare professional (PLHCP) is defined as‘‘an individual whose legally permittedscope of practice (i.e., license,registration, or certification) allows himor her to independently provide, or bedelegated the responsibility to provide,some or all of the health care servicesrequired by paragraph (e) of thissection.’’ This definition has been addedbecause paragraph (e)(2) of the finalstandard requires that all medical

evaluation procedures be performed bya PLHCP.

OSHA has long considered the issueof whether, and if so how, to specify thequalifications of the particularprofessionals who are permitted toperform the medical evaluationsrequired by its standards. The Agencyhas determined that any professionalwho is licensed by state law to performthe medical evaluation proceduresrequired by the standard may performthese procedures under the respiratoryprotection standard. The Agencyrecognizes that this means that thepersonnel qualified to provide therequired medical evaluation may varyfrom state to state, depending on statelicensing laws. Under the final rule, anemployer has the flexibility to retain theservices of a variety of qualifiedlicensed health care professionals,provided that these individuals arelicensed to perform a given service.OSHA believes that this flexibility willreduce cost and compliance burdens foremployers and increase convenience foremployees. The approach taken in thisfinal standard is consistent with theapproach OSHA has taken in otherrecent standards (e.g., cadmium,methylene chloride).

Positive pressure respirator. This termhas been redefined in the final standardto mean ‘‘a respirator in which thepressure inside the respiratory inletcovering is positive with respect toambient air pressure outside therespirator.’’ Consistent with therecommendations of severalcommenters (Exs. 54–208, 54–218, 54–219, 54–410, and 54–424), the finalstandard’s definition adopts the ANSIZ88.2–1992 definition but adds thephrase ‘‘outside the respirator’’ forclarity.

Powered air-purifying respirator. Thefinal standard defines this term as ‘‘anair-purifying respirator that uses ablower to force the ambient air throughair-purifying elements to the inletcovering.’’ This revision also reflectscommenters’ recommendations thatOSHA adopt ANSI Z88.2–1992 standarddefinitions (Exs. 54–208, 54–218, 54–219, 54–410, and 54–424). The term‘‘ambient atmosphere’’ in the ANSIdefinition has been replaced with theterm ‘‘ambient air’’ for simplicity.

Pressure demand respirator. This typeof respirator is defined as ‘‘a positivepressure atmosphere-supplyingrespirator that admits breathing air tothe facepiece when the positive pressureis reduced inside the facepiece byinhalation.’’ This language has beentaken verbatim from the ANSI Z88.2–1992 standard’s definition, except that


the term ‘‘breathing air’’ has replacedthe term ‘‘respirable gas’’ for clarity.

Qualitative fit test (QLFT). Thisdefinition has been revised to read ‘‘apass/fail fit test to assess the adequacyof respirator fit that relies on theindividual’s response to the test agent.’’OSHA has replaced the proposal’s QLFTdefinition with one derived from theANSI Z88.2–1992 standard but hasadded the phrase ‘‘to assess theadequacy of respirator fit’’ to emphasizethe purpose of QLFT. In addition, theOSHA definition uses the phrase ‘‘theindividual’s response’’ instead of theANSI definition’s phrase ‘‘subject’ssensory response’’ for clarity.

Quantitative fit test (QNFT). Thisdefinition has been revised andsimplified to accommodate both currentand yet-to-be-developed fit testtechnology. The final standard defines aquantitative fit test (QNFT) as ‘‘anassessment of the adequacy of respiratorfit by numerically measuring theamount of leakage into the respirator.’’Commenters generally opposed theproposed definition of QNFT, whichmade reference to challenge agents,because they feared that it mightinterfere with the development of newfit test methods (Exs. 54–5, 54–222, 54–251, 54–266, 54–275x, 54–350, 54–208,54–218, 54–219, 54–278, 54–316, 54–424). OSHA agrees and has revised thedefinition accordingly. OSHA believesthat the definition of QNFT must beusable, enforceable, andunderstandable, and accommodateevolving technology.

Respiratory inlet covering. The finalstandard defines this term, which isoften used in descriptions of respiratoryequipment, as ‘‘that portion of arespirator that forms the protectivebarrier between the user’s respiratorytract and an air-purifying device orbreathing air source, or both. It may bea facepiece, helmet, hood, suit, or amouthpiece respirator with noseclamp.’’ This definition is adapted fromthat in the ANSI Z88.2–1992 standard;the phrase ‘‘that connects the wearer’srespiratory tract’’ in the ANSI definitionhas been modified to read ‘‘that formsthe protective barrier between the user’srespiratory tract’’ in the OSHAdefinition for clarity.

Self-contained breathing apparatus(SCBA). The proposed definition of self-contained breathing apparatus (SCBA)has been revised slightly in the finalstandard to read ‘‘an atmosphere-supplying respirator for which thebreathing air source is designed to becarried by the user.’’ This reviseddefinition was adopted from the ANSIZ88.2–1992 standard’s definition ofSCBA.

Service life. The final standard definesservice life as ‘‘the period of time thata respirator, filter, or sorbent, or otherrespiratory equipment providesadequate protection to the wearer.’’ Thisdefinition eliminates a reference in theproposal to substances ‘‘breakingthrough’’ the cartridge or canister, anddeletes a statement that respiratormanufacturers are to determine servicelife concentrations, since this is theemployer’s responsibility. The newdefinition parallels ANSI’s except that itcontains additional language coveringfilters, sorbents, and other respiratoryequipment. This definition is furtherexplained in the discussion ofparagraph (d) of the Summary andExplanation.

Supplied-air respirator (SAR) orairline respirator. OSHA has elected toretain a definition for supplied-airrespirators, since the term is used byNIOSH in the 42 CFR part 84regulations. The final standard’sdefinition reads: ‘‘Supplied-airrespirator (SAR) or airline respiratormeans an atmosphere-supplyingrespirator for which the source ofbreathing air is not designed to becarried by the user.’’ Participants (Exs.54–208, 54–249) were more familiarwith this term than with the term ‘‘air-supplied respirator’’ recommended asan alternative by some commenters(Exs. 54–218, 54–219, 54–363, 54–434).The language of this definition isderived from the ANSI Z88.2–1992definition for ‘‘airline respirator,’’ butalso applies to supplied-air respirators,a term that NIOSH uses to certify thisclass of respirators. OSHA believes thatusing both names in the definition willreduce confusion for respirator users.

Tight-fitting facepiece is defined as ‘‘arespiratory inlet covering that forms acomplete seal with the face.’’ This termwas not defined in the proposal, butnumerous commenters requested thatOSHA add this definition (Exs. 54–222,54–283, 54–363, 54–410, 54–424, 54–428, 54–433, 54–455) to the finalstandard.

User seal check is defined as ‘‘anaction conducted by the respirator userto determine if the respirator is properlyseated to the face.’’ Such a check isperformed by the user each time therespirator is donned or adjusted toensure that the tight-fitting respirator isproperly seated on the user’s face, i.e.,that the proper seal has been achieved.Several commenters recommended thatOSHA add the definition for ‘‘fit check’’from the ANSI Z88.2–1992 standard toreplace the term ‘‘facepiece seal check’’that was used in Appendix B of theproposal (Exs. 54–208, 54–218, 54–219,54–410, 54–424). The term ‘‘fit check’’

has proven confusing to those respiratorusers who do not realize that a daily fitcheck is not a substitute for an annualfit test. The AIHA (Ex. 54–208)recommended that OSHA add astatement to Appendix B to the effectthat: ‘‘Fit checks are not substitutes forqualitative or quantitative fit tests,’’ andOSHA has done so in this finalstandard. Because OSHA believes thatthe similarity between the terms ‘‘fitcheck’’ and ‘‘fit test’’ is responsible forthis confusion, OSHA has used the term‘‘user seal check’’ rather than ‘‘fitcheck’’ in the final standard. Thedefinition of ‘‘user seal check’’ derivesfrom the ANSI Z88.2-1992 standard’sdefinition for ‘‘fit check,’’ except thatthe word ‘‘action’’ has been substitutedfor ‘‘test’’ to avoid any possibleconfusion among respirator users.

Paragraph (c)—Respiratory ProtectionProgram

This paragraph of the final standardrequires employers to develop andimplement a written respiratoryprotection program, with workplace-specific procedures addressing themajor elements of the program,whenever respirators are necessary toprotect the health of the employee. Inaddition, where an employer requires anemployee to wear a respirator, i.e., in asituation where the standard does nototherwise require such use, a writtenprogram must be developed andimplemented. Employers who providerespirators at the request of theiremployees or who allow theiremployees to bring their own respiratorsinto the workplace must ensure that therespirator used does not present ahazard to the health of the employee.However, if the respirator voluntarilyworn is a filtering facepiece (dust mask),the employer is not required toimplement a written program. Paragraph(c)(1) also requires employers to updatethe program when changes in theworkplace or in respirator use makesuch updating necessary.

As in the proposed rule, the finalstandard requires that the respiratoryprotection program be written. OSHA’sexperience and that of the industrialhygiene community have demonstratedthat health and safety programs can bestbe effectively implemented andevaluated when written. In addition,because workplaces differ substantially,each program must be tailored to thespecific conditions of the workplace ifit is to protect employee health, anddeveloping a written program is themost efficient way of ensuring that theprogram reflects the uniquecharacteristics of each workplace.Developing and writing down worksite-


specific procedures requires employersto design their respiratory protectionprograms to address the respiratoryhazards in their particular workplace,and this process requires employers tothink about and document all relevantinformation pertaining to the hazardousatmospheres that their employees mayencounter under normal operatingconditions or during reasonablyforeseeable emergencies that may occurin the workplace. Finally, OSHA’senforcement data indicate thatcompliance with the previous standardhas not been optimal, particularly insmaller workplaces, and a writtenprogram will help employers,employees, and compliance officersgauge the adequacy of a given program.

Paragraphs (c)(1)(i) through (c)(1)(ix)identify the elements that must beincluded in the employer’s programunless the particular element does notapply to the employer’s workplace. Theprevious OSHA respiratory protectionstandard also required employers todevelop written standard operatingprocedures that covered the selection,use, cleaning, maintenance, inspection,and storage of respirators and thetraining and medical evaluation ofrespirator users (paragraphs (b)(1),(e)(1), and (e)(3), among otherprovisions of the previous standard). Inthe final standard, the general elementsof the written program have beenexpanded, reordered and updated, andthe term ‘‘written standard operatingprocedures (SOP)’’ used in the previousstandard has been replaced with thewords ‘‘worksite-specific procedures.’’Thus, the standard identifies the basicelements of written programs for allworkplaces, but the employer has theflexibility to tailor these generalprogram elements to match the specificworkplace conditions and processes thatoccur in that workplace. In the Agency’sprevious respiratory protectionstandard, the requirement for writtenstandard operating procedures tended tolead to the adoption of genericprocedures. Changing the terminologyfrom ‘‘SOPs’’ to ‘‘worksite-specificprocedures’’ gives employers theincentive to develop procedures that areunique and specific to the employer’sworkplace, to describe the particularrespirator selection process used in thatworkplace, and to explain howemployees are to use respirators in thatsetting.

OSHA has also revised the requiredprogram elements themselves, forseveral reasons. First, they have beenmodified to reflect those provisions ofthe final standard that have been addedor enhanced to reflect advances inrespiratory protection technology, such

as the development of atmosphere-supplying respirators and thewidespread use of modern methods offit testing. Second, several of theprovisions of the previous standardwere vague and had caused compliancedifficulties for employers over the years.OSHA wishes to provide employerswith clear notice of what elementsOSHA considers essential to an effectiverespirator program. Third, OSHA hasadopted several changes suggested bycommenters.

OSHA also believes that clearerprogram elements will improveemployer compliance. According to theMinnesota Department of Labor andIndustry (Ex. 54–204), for example,many employers have had difficultycomplying with OSHA’s previousstandard because they were unsure whatelements a program was required toinclude. Several other data sources alsopoint to the lack of clarity in OSHA’sprevious standard; these includeOSHA’s inspection data and complianceexperience, comments to the record (Ex.54–219), and studies of workers (Ex. 64–65). As noted in the NPRM, datacollected on current respirator practicesand procedures in over 2300manufacturing plants classified in 15SIC codes were reviewed by the Agency(See Summary of the PreliminaryRegulatory Impact Analysis, 59 FR58892). This survey sample was used toproduce estimates of respirator-relatedpractices for about 123,200manufacturing plants with regular andoccasional respirator use. Only 25.5% ofthese plants were estimated to havewritten standard operating procedures,and only 7.9% had procedures thataddressed all eight of the programelements required by the previousstandard (selection, use, cleaning,maintenance, inspection and storage ofrespirators, and the training andmedical evaluation of respirator users).More than 80% of the very large plants(those with 1000 or more employees)had written procedures, while in smallplants (those with fewer than 50employees), only about 22% had writtenprocedures. This survey clearly showedthat improving the clarity of theelements to be addressed in standardoperating procedures would helpemployers to develop and implementbetter respiratory protection programsand thus would provide greaterprotection to workers as well.

Similarly, a study of OSHA citationsfor violations of the previous OSHArespirator standard from 1977 to 1982showed that 13% of these citations wereissued because standard operatingprocedures were either inadequate ormissing (Rosenthal and Paull; Ex. 33–5).

OSHA’s latest citation data for therespiratory protection standard, for theperiod October 1990 to December 1995,show that the number of citations issuedfor inadequate or missing writtenrespirator programs in general industryhas increased to 18.4% of all respiratorstandard-related citations. These dataindicate that the conclusions reached byRosenthal and Paull are still valid. Thecitation history for the constructionindustry respiratory protectionstandard, 29 CFR 1926.103, is similar,with citations for inadequate respiratorprograms representing 10.5% of allrespirator standard-related citations inthat industry. OSHA believes that thepercentages of respirator standard-related citations reported in thesereviews substantially underestimate thereal incidence of deficient programsbecause it is OSHA policy not to issuecitations for an inadequate programunless an overexposure is alsodocumented.

Paragraphs (c)(1)(i) through (c)(1)(ix)of the final standard provide additionaldetail about each of the requiredprogram elements but remainperformance based to enable employersto adapt them to their workplaces. Theprogram elements have beenreorganized from those in the previousstandard so that they track the order ofthe major paragraphs of the standard.OSHA believes that reordering theelements, as suggested by onecommenter (Ex. 54–204), is logical andshould make program developmenteasier. OSHA also believes that theadditional detail and greater clarityprovided by the final rule’s programelements will reduce confusion over theintent of these provisions, lead to highercompliance rates, and result in betterrespiratory protection for employees.

The ANSI Z88.2–1992 standard forrespiratory protection also states thatwritten procedures covering thecomplete respirator program must beestablished and implemented (Ex. 81).Thus, like OSHA, ANSI recognizes theneed for a written respiratory protectionprogram and implementing proceduresto provide complete and consistentprotection to employees wearingrespirators. Although the ANSI standarddoes not contain detailed instructionson the content of these procedures, itdoes describe, in clause 6, the elementsto be included in the program to coverroutine and emergency use ofrespirators.

The program elements in the ANSIZ88.2–1992 standard (i.e., programadministration, respirator selection,training, respirator fit, maintenance,inspection and storage) are similar tothose in paragraphs (c)(1)(i) through


(c)(1)(ix) of OSHA’s final standard. Thespecific content of each element of thewritten procedures is left to theemployer, who can tailor them to matchthe conditions that occur in his/herworksite. Although many of the programelements are common to all respiratoryprotection programs, such as respiratorselection, care, use, and programevaluation, some elements, such as theone addressing specifications for airquality for atmosphere-supplyingrespirators, apply only in workplaces inwhich those types of respirator are used.

OSHA received many comments, bothon written programs in general and onspecific program elements. Somecommenters (Exs. 54–160, 54–187, 54–238), questioned the need for a writtenrespirator program with worksite-specific procedures. For example,Transtar Railroads (Ex. 54–160) statedthat written procedures do notguarantee an effective respiratoryprotection program and argued thatrequiring additional written programelements would not cause thosecompanies who presently disregardOSHA’s existing standard to becomemore conscientious. Motorola (Ex. 54–187) urged OSHA to delete therequirement for a written program andinstead simply to require that employersensure that respirators are properlyselected, fitted, used, and maintained asnecessary to protect employees whenrespirators are required. However, therequirement for a written respiratorprogram was widely supported by manyother participants in the rulemaking(Exs. 54–204, 54–219, 54–304, 54–387,54–389, 54–428, 54–435). For example,the United Automobile Workers (Ex.54–387) agreed that a written respiratoryprotection program that is site-specificand detailed (for example, that includesspecific procedures for determiningwhen a cartridge or filter needs to bechanged) should be required. TheAmerican Federation of Labor andCongress of Industrial Organizations(AFL–CIO) (Ex. 54–428) stronglysupported the requirement for a writtenrespiratory program and identified sucha program as the fundamental core ofthe standard:

The AFL–CIO strongly supports theAgency’s proposal that employers who arerequired to use respirators or voluntarily userespirators in the workplace establish awritten respiratory protection program. Thewritten program constitutes an employer’splan for dealing with worker protection fromhazardous airborne contaminants that may bepresent in the workplace, and as such, weview these provisions as the fundamentalcore of the standard. Requiring a writtenprogram is essential in providing uniformityand consistency while supplying the

maximum protection for workers who userespirators in the workplace. (Ex. 54–428)

OSHA’s expert witness, JamesJohnson of the Lawrence LivermoreNational Laboratory, testified thatrespiratory protection programs must bewritten because of their complexity:

* * * A respirator program involves manydecisions. What kind of respirator do I use,what kind of concentrations were measured,what kind of contaminants were in theworkplace

* * * So all this information is importantto provide documentation and understandingso that you can make sure the program isadequate and you can make changes to it, toimprove it and to have it be a dynamicoperation as the workplace changes * * *(Tr. 212)

Commenting in the same vein, theNational Pest Control Association (Ex.54–435), which represents many smallbusinesses, agreed that requiringemployers to provide a writtenrespiratory program was sensible, andthe Cambrex Corporation (Ex. 54–389)noted that ‘‘A performance approach indefining written program requirementswill provide needed flexibility toemployee protection programs.’’ DavidLee, CIH, CSP (Ex. 54–304), stronglysupported the approach OSHA hastaken in the final rule; he stated that awritten respiratory protection programshould be required in all places whererespirators are used, regardless of thecircumstances, and that the program’scontents should be specifically tailoredto conditions of use at the place ofemployment.

OSHA agrees with these commentersthat it is appropriate to retain theprevious standard’s requirement for awritten program, and that the programmust be flexibly tailored to worksiteconditions. OSHA finds that commentsto the record, and the Agency’s owncompliance experience, strongly suggestthat many employers wish to complybut are unsure about what is required;for these employers, greater clarity andguidance will enhance compliance andenable them to provide their employeeswith needed protection.

Paragraph (c)(1) of the final rulerequires employers to update theprogram as necessary to reflect changesin the workplace. This requirement hasbeen revised somewhat from theproposal. The proposed standard statedthat ‘‘[t]he written program shall reflectcurrent workplace conditions andrespirator use’’ (59 FR 58939). OSHAreceived several comments on thisprovision (Exs. 54–278, 54–213, 54–249). For example, the Dow ChemicalCompany (Ex. 54–278) urged OSHA torevise this language to require that theprogram reflect only those current

workplace conditions ‘‘significantlyimpacting respirator use.’’ In the finalrule, OSHA has moved this provision toparagraph (c)(1) and revised it to requirethat the program be ‘‘updated asnecessary to reflect those changes inworkplace conditions that affectrespirator use.’’ OSHA believes that thischange is responsive to Dow’s point. Asnow written, when the workplacechanges in a way that may affectrespirator use, such as when newprocesses are introduced, changes aremade in the types of chemicals used, orthe types of respirators being usedchanges, employers must revise theprogram as necessary to reflect thesenew conditions.

One of the major issues raised in therulemaking dealt with situations inwhich respirator use is not specificallyrequired by 29 CFR 1910.134 or otherOSHA statutory or regulatoryrequirements, but instead is required byemployers as a condition ofemployment or is permitted byemployers upon the request ofemployees (i.e., voluntary use). Thepreamble discussion for proposedparagraph (a) stated that employers whorequired employees to use respiratorswould be covered by the standard (59FR 58895). OSHA also recommended inthe NPRM that employers who permitvoluntary respirator use in theirworkplaces implement the fullrespiratory protection program. In thefinal rule, paragraph (c)(1) requires thata respiratory protection program bedeveloped and implemented ‘‘whereverrespirators are required by theemployer,’’ but has greatly reduced theobligations of employers who allowtheir employees to use respirators whensuch use is not required.

In the preamble to the proposal,OSHA discussed the reasoning behindincluding employer-required respiratoruse within the scope of the standard (59FR 58895). OSHA stated that therequirement was appropriate bothbecause the use of a respirator could initself present a health hazard to thewearer, and because improper use of arespirator in environments whererespiratory hazards are present wouldnot sufficiently protect employees fromthose hazards. OSHA finds that theseare still valid reasons for requiring thata respiratory protection program beimplemented where employers requirerespirator use. All of the elements of arespiratory protection program apply tothis situation. Employers must stillselect respirators that are appropriate tothe workplace conditions and types ofrespiratory hazards present to ensurethat respirators offer adequateprotection. Improperly selected


respirators may afford no protection atall (for example, use of a dust maskagainst airborne vapors), may be souncomfortable as to be intolerable to thewearer, or may hinder vision,communication, hearing, or movementand thus pose a risk to the wearer’ssafety or health.

Employees who are required by theiremployers to wear respirators must alsobe medically evaluated to determinethat they are capable of tolerating theincreased physiological load associatedwith some respirator use. Proper fittesting is necessary to ensure thatdiscomfort is minimized and that therespirator selected is offering sufficientprotection. It is also necessary thatrespirators required by employers becleaned, disinfected, stored, inspected,and repaired according to theprocedures contained in the final rule toensure proper respirator functioningand protection of employees fromdermatitis or exposure to hazardouscontaminants that may result from usinga dirty respirator. Compliance with theprovisions of the standard dealing withsupplied air quality and use is alsoessential where employers require theuse of supplied-air respirators. Whenemployers require employees to userespirators, OSHA believes it necessarythat employees be properly trained intheir use and care, and be informed ofthe limitations of using respirators.Paragraph (k) of the final rule makesclear that employers must implementthe employee training requirementscontained in paragraph (k) if theyrequire their employees to userespirators.

In contrast, not all of these protectionsare necessary in the situation where anemployer allows, but does not require,respirator use. OSHA has thereforeadded a new paragraph (c)(2) to the finalrule, which applies when employersallow employees to use respiratorswhen such use is not required by theemployer or by the standard. Thisparagraph applies when employerseither provide respirators to employeeswho request them or allow employees touse their own respirators. In bothsituations, paragraph (c)(2)(i) states thatemployers must determine that theemployees that they allow to userespirators are medically able to do so,and that there are no other conditionsthat could cause the respirator use tocreate a hazard.

If the employer allows voluntaryrespirator use, paragraph (c)(2)(i)requires that the employer provide theemployee with the informationcontained in Appendix D to thisstandard, entitled ‘‘Information forEmployees Using Respirators When Not

Required Under the Standard.’’ In therare case where an employee isvoluntarily using other than a filteringfacepiece (dust mask) respirator(paragraph (c)(2)(ii)), the employer mustimplement some of the elements of arespiratory protection program, e.g., themedical evaluation component of theprogram and, if the respirator is to bereworn, the cleaning, maintenance, andstorage components. An exception tothis paragraph makes clear that, wherevoluntary respirator use involves onlyfiltering facepieces (dust masks), theemployer is not required to implementa written program.

Paragraph (c)(2) is necessary becausethe use of respirators may itself presenta health hazard to employees who arenot medically able to wear them, whodo not have adequate information to useand care for respirators properly, andwho do not understand the limitationsof respirators. Paragraph (c)(2) isintended to allow employers flexibilityto permit employees to use respiratorsin situations where the employees wishto do so, without imposing the burdenof implementing an entire respiratorprogram. At the same time, it will helpensure that such use does not create anadditional hazard and that employeesare provided with enough informationto use and care for their respiratorsproperly. This provision does not, ofcourse, preclude employers fromadopting additional program elements ifthey believe such elements areappropriate.

The great majority of voluntary usesituations involve the use of dust masks,i.e., filtering facepieces, which areprovided for the employee’s comfort.For example, some employees who haveseasonal allergies may request a maskfor comfort when working outdoors, oran employee may request a dust maskfor use while sweeping a dusty floor.There are no medical limitations on theuse of these respirators, so employerswho allow their use need only ensurethat the masks are not dirty orcontaminated, that their use does notinterfere with employees’ ability towork safely, and that they provide theemployees with the informationcontained in Appendix D, as requiredby paragraph (k) of the final rule.

In rare cases where the employeerequests and the employer allows theuse of a negative-pressure respirator(tight-fitting), or where the employeebrings such a respirator into theworkplace, the employer mustimplement some provisions of therespirator program described inparagraph (c)(1) to ensure that suchrespirator use will not affect theemployee’s health adversely. The

employer can include these elements inits existing respiratory protectionprogram, if it is required to maintainone. Some medical evaluation isnecessary to determine that theemployee is physically able to use atight-fitting negative pressure respirator.In addition, if the respirators being usedvoluntarily are reused, it is necessary toensure that they are maintained inproper condition to ensure that theemployee is not exposed to anycontaminants that may be present in thefacepiece, and to prevent skin irritationand dermatitis associated with the useof a respirator that has not been cleanedor disinfected. OSHA believes itunlikely that voluntary use situationswill involve the use of supplied-airdevices, but such use would also triggerthese requirements of the standard.

These requirements are necessarybecause use of a negative pressure(tight-fitting) respirator imposes asignificant physiologic burden on arespirator user, and it is crucial todetermine that the user can withstandthat burden without suffering adversehealth consequences. Similarly,reusable tight-fitting negative pressurerespirators can become contaminated ifthey are not cleaned, maintained, andstored properly. Thus if an employerallows use of this type of respirator, theemployer must implement the programelements necessary to ensure thatcontamination does not harm theemployee.

The hazards addressed by thisrequirement are the same ones that arealready considered under OSHA’slongstanding enforcement policy. TheAgency generally does not issuecitations for violations of its respiratorstandards unless there is also evidenceof overexposure to a hazardoussubstance, or some other hazard causedby improper or inadequate respiratoruse. (OSHA Field Inspection ReferenceManual (FIRM), Ch. III. Sec. C.3.c).Other hazards referenced in the FIRMinclude ingestion of harmful substancesthat may remain on improperly cleanedand maintained respirators, ordermatitis caused by the samecondition. These are precisely thehazards that the requirements ofparagraph (c)(2) are designed to prevent.They can occur whether respirator useis voluntary or required, and OSHAdoes not believe it would be consistentwith the OSH Act to allow employees toexpose themselves to preventablehazards, particularly where there arefairly undemanding measures availableto prevent that exposure.

Requiring employers to undertakethese minimal obligations when theyallow voluntary respirator use is


consistent with the fact that employerscontrol the working conditions ofemployees and are therefore responsiblefor developing procedures designed toprotect the health and safety of theemployees. Employers routinelydevelop and enforce rules andrequirements for employees to followbased on considerations of safety. Forexample, although an employer allowsemployees discretion in the types ofclothing that may be worn on site, theemployer would prohibit the wearing ofloose clothing in areas where clothingcould get caught in machinery, orprohibit the use of sleeveless shirtswhere there is a potential for skincontact with hazardous materials.Similarly, if an employer determinesthat improper or inappropriaterespirator use presents a hazard to thewearer, OSHA finds that the employermust exert control over such respiratoruse and take steps to see that respiratorsare safely used under an appropriateprogram. It has been OSHA’s experiencethat employers will be able to determinewhether employees are using their ownrespirators in the workplace, just as theyare able to determine that employees areadhering to all other procedures andrequirements established by theemployer.

Concomitantly, OSHA’s decision toimpose fewer requirements on voluntaryrespirator use than on required use issupported by the record. Manycomments addressed the issue of howthe final standard should treat these twotypes of respirator use. Manycommenters (Exs. 54–96, 54–109, 54–196, 54–222, 54–272, 54–341, 54–424,145, 176, Tr. 2127, Tr. 2174 ) supportedthe inclusion of employer-requiredrespirator use, but not of voluntary use,within the full scope of the standard.Many of these rulemaking participantsbelieved that voluntary respirator useshould require a minimal programdesigned to provide information andtraining to the employee, and that otherelements of the program should not bemade mandatory. Typical of these wasthe post-hearing comment ofOrganization Resources Counselors, Inc.(ORC):

OSHA should not require a completerespirator program for the voluntary use ofrespirators by employees, when not requiredby an OSHA standard, or by the employer.Some employees will wish to use respiratorseven though they are not required to protectagainst overexposure to a toxic hazard. Inthese instances the employer should berequired only to inform the employee of thesafe and proper use of such respirators andany associated limitations on the particulardevice chosen (Ex. 145).

In addition, some of these commenters(Exs. 54–341, 176, Tr. 594, Tr. 2100)suggested that requiring employers tocomply with all or most of therequirements would discourageemployers from permitting voluntaryrespirator use in their workplaces. Forexample, in its post-hearing submission,the North American InsulationManufacturers Association (NAIMA)commented as follows:

NAIMA agrees with many other hearingparticipants that employers should berequired to train voluntary respirator users inthe proper function and use of respirators* * * OSHA should, however, tailor otheraspects of the Proposed Rule to ensure thatthe more onerous and unnecessary additionalrequirements, such as comprehensivemedical examinations, are not imposed intruly voluntary use situations. Applyingunnecessary ancillary requirements tovoluntary use situations would discourageemployers from allowing workers such use(Ex. 176).

OSHA believes that the final ruleprovides for the kind of tailoringsuggested by NAIMA’s comment.Employers who permit the voluntaryuse of tight-fitting negative-pressurerespirators must utilize the proceduresnecessary to address the health hazardsassociated with the use of suchrespirators, but in the vast majority ofvoluntary-use situations whereemployees are using dust masks(filtering facepieces), the standard doesnot require the employer to implementa written respirator program to ensureemployee health. Thus, the final ruledoes not require employers providingdust masks (filtering facepieces) to theiremployees to comply with therequirements that NAIMA considers‘‘onerous and unnecessary’’ in thissituation. However, where respiratorsare used voluntarily by employees, andthe use of a given type of respirator, e.g.,a tight-fitting negative pressurerespirator, is associated with anincreased health risk, OSHA finds thatapplying relevant portions of therespiratory protection program isessential to ensure worker protection.

Other commenters (Exs. 54–214, 54–218, 54–278, 54–389) believed thatapplication of the standard should belimited in situations where there was noexposure to a respiratory hazard,regardless of whether respirator use isrequired by employers in this situationor is voluntary. In discussing this issue,the 3M Company commented asfollows:

1. Any use of respirators or masks in theworkplace should trigger a requirement for atleast a minimal respiratory protectionprogram. Regardless of whether use isrequired or recommended by an employer or

is self-imposed by an employee, theemployer should be responsible for the safeuse of respirators and masks in theworkplace.

2. Where it is documented by an employerthat no hazard exists—such as when usedagainst non-toxic materials, exposures wellbelow the permissible exposure limit (PEL)or hazard level, or voluntary use against suchconditions as discomfort or allergies—therule should only require an abbreviatedrespiratory protection program * * *. (Ex.54–218)

In a similar argument, the DowChemical Company (Ex. 54–278)suggested that employers be exemptfrom the standard’s requirements if theyrequire employees to use respirators asa precautionary measure whereexposures are below the PELs.

OSHA did not adopt this approach inthe final rule because the Agencybelieves that, in most cases of employer-required respirator use, respirators arebeing used as protection against actualor potential exposure to a respiratoryhazard. In these cases, OSHA finds thatit is necessary and appropriate that theemployer implement all elements of therespiratory protection program thatapply to the worksite-specificconditions under which respirators areused. If respirators are used asprotection against a real or potential riskcaused by exposure to a respiratoryhazard, OSHA believes it essential forthe employer to provide for properrespirator selection, fit testing, medicalevaluation, and care and maintenance toensure that the respirator is providingsufficient protection against the hazardand that use of the respirator is notimposing an additional health risk.OSHA also believes that, bydistinguishing between employer-required and voluntary respirator use inthe final rule, it will be easier foremployers to determine the extent towhich the standard will apply to theirspecific workplaces.

Other rulemaking participants (Exs.54–208, 177, Tr. 782, Tr. 1722) were ofthe opinion that voluntary respirator useshould not be distinguished fromemployer-required use in determininghow the standard should apply, orreported that some employers alreadyimplement a program for voluntary use.The AIHA, in support of full coverageof the standard for voluntary respiratoruse, stated in written comment:

The position of AIHA is that all use ofrespiratory protection should be covered byan employer’s respiratory protectionprogram. That includes both voluntary use aswell as required use. Both groups shouldparticipate in all elements of the respiratoryprotection program. An individual desiringto wear a respirator to obtain some level ofcomfort or to further reduce their exposure to


a chemical in the workplace should receivethe full benefits of an established program:training to convey proper knowledge inequipment selection, maintenance, and use;medical evaluation to confirm that its usewill not present a risk to the individual; andfit testing to confirm that the equipment fitsproperly and workplace surveillance toconfirm that the equipment being utilized issuitable for the exposure level. (Ex. 54–208)

At the public hearing, Larry Janssen ofthe AIHA elaborated that ‘‘* * * thereshould be some kind of a minimumframework to prevent the misuse ofrespirators in those voluntary usesituations, that you don’t do harm byallowing a respirator to be used whereit’s not really needed’’ (Tr. 782).Similarly, in a post-hearing comment,the Industrial Safety EquipmentAssociation (ISEA) stated that it wasimportant to cover voluntary use in thestandard since ‘‘* * * [r]espirators thatare not used properly could present ahazard’’ (Ex. 177). This practice isalready being implemented in someworkplaces; Richard Holmes of UnionCarbide, representing the ChemicalManufacturers Association (CMA) at thehearings (Tr. 1722), testified that ‘‘* * *[w]e treat the voluntary user just like amandatory user so they’re in theprogram just as though they wererequired to wear the respirator and the* * * medical surveillance is allhandled the same * * * [as is thetraining].’’

As discussed above, OSHA agrees thatsome voluntary respirator use (e.g., thatinvolving tight-fitting negative-pressurerespirators) may present a health hazardto employees if the respirator is notproperly selected, maintained, andused. Therefore, OSHA has revised thefinal rule to ensure that employers whopermit voluntary use of such respiratorsin their workplaces implement thoseportions of the standard necessary toprotect employees from any health risksassociated with respirator use. Theposition taken in the final rule alsoreflects OSHA’s long-standingenforcement policy with the previousrespiratory protection standard, asstated in the FIRM and in several lettersof interpretation issued by the Agency(See letters dated 10/2/87 from ThomasJ. Shepich, 4/11/91 from Patricia K.Clark, 3/19/91 from Patricia K. Clark, 3/4/93 from Roger A. Clark (2 letters), and3/15/95 from Ruth McCully). Forexample, in the letter of March 4, 1993from Roger A. Clark, OSHA stated itspolicy regarding the application of 29CFR 1910.134 to the voluntary use ofrespirators:

OSHA’s policy is that if the respirator itselfcould present an adverse health condition ifa specific requirement of the respiratory

protection standard is not observed, then therequirement applies. Examples may includea dirty respirator that is causing dermatitis,a worker’s health being jeopardized bywearing a respirator due to an inadequatelyevaluated medical condition, or a significantingestion hazard created by an improperlycleaned respirator. This is so regardless ofwhether the employee purchased therespirator or the employer provides it.

OSHA also has determined thatcomplete training is not required foremployees using respirators voluntarily.Instead, paragraph (k) of the final rulerequires employers to provide theinformation contained in Appendix D toensure that employees are informed ofproper respirator use and the limitationsof respirators.

Paragraphs (c)(1)(i) through (c)(1)(ix)list the elements of the respiratorprogram required by this standard.Paragraph (c)(1)(i) requires the programto contain procedures for the selectionof respirators appropriate to protectemployees from the respiratory hazardspresent in the particular workplace.This provision is unchanged from thecorresponding provision in the proposaland is also similar to paragraph (b)(2) ofOSHA’s previous standard. Paragraph(c)(1)(ii) addresses the medicalevaluation of employees required towear respirators and is unchanged fromthe parallel requirement in the proposal.The AIHA (Ex. 54–208) recommendedthat paragraph (c)(1)(ii), which requiresemployers to develop proceduresaddressing ‘‘medical evaluations ofemployees required to wearrespirators,’’ be changed to specify thatthese procedures need only coveremployees who are ‘‘authorized by theemployer to wear respirators’’; the AIHAwanted this word change to ensure thatemployers understood that theseprocedures must cover both voluntaryand required use. However, asexplained above, OSHA has decided torequire medical evaluation of employeeswho use respirators voluntarily onlywhen such use may present a healthhazard to employees, e.g., in the case oftight-fitting negative pressurerespirators. Therefore, OSHA has notincluded the language suggested by theAIHA in the final rule.

Paragraph (c)(1)(iii) covers the fit testelement of the program and has beenmodified since the proposal to respondto comments. The proposal would haverequired the program to contain fittesting procedures ‘‘for air-purifyingrespirators and tight-fitting positivepressure respirators.’’ The ServiceEmployees International Union (Ex. 54–455) commented that this provisiononly needed to address ‘‘tight-fittingrespirators’’ because this language

adequately describes the respiratoryequipment to be covered. Since OSHAhas revised the fit testing requirementsin paragraph (f) to cover all tight-fittingrespirators, the language in paragraph(c)(1)(iii) has been revised accordingly.

Paragraph (c)(1)(iv) states thatemployers shall include ‘‘Procedures forproper use of respirators in routine andreasonably foreseeable emergencysituations.’’ In the NPRM, thisrequirement was addressed underparagraph (g)(1), but it has been movedinto paragraph (c)(1) of the final rule toensure that employers are aware thatwritten workplace-specific proceduresmust address both routine and non-routine respirator usage, including thatin reasonably foreseeable emergencysituations. OSHA received no commentson this provision.

Paragraph (c)(1)(v) requires theworkplace-specific procedures to cover‘‘procedures and schedules for cleaning,disinfecting, storing, inspecting,repairing, discarding, and otherwisemaintaining respirators.’’ This provisionis unchanged from that proposed. TheAmerican Iron and Steel Institute (AISI)urged OSHA to remove the word‘‘schedules’’ from paragraph (c)(1)(iv)and to substitute the word‘‘frequencies’’ instead. AISI stated thatthe term ‘‘schedules’’ connotes arequirement for extensive recordkeepingand paperwork. OSHA does not agree.Since OSHA requires the respiratorprogram to be written, as required underthe prior standard and as proposed andsupported by comments in thisrulemaking, it is OSHA’s conclusionthat including the employer’s schedulefor cleaning, disinfecting, or otherwisemaintaining respirators is not undulyburdensome. A schedule is needed toinform employees when they are to havetheir respirators fit tested, cleaned, andmaintained. Therefore, OSHA isretaining the word ‘‘schedule.’’Representatives of the ServiceEmployees International Union [(SEIU)Ex. 54–455)] strongly supported therequirement for maintenance schedulesas proposed under paragraph (c)(1)(v) ofthe NPRM for the same reason.

Paragraph (c)(1)(vi) is essentiallyunchanged from the proposal andrequires ‘‘Procedures to ensure adequateair quality, quantity, and flow ofbreathing air for atmosphere-supplyingrespirators.’’ Representatives from SEIU(Ex. 54–455) supported OSHA’saddition of ‘‘quantity and flow’’ toparagraph (c)(1)(vi) in the NPRM. Properair quality and quantity are crucial tothe use of supplied air respirators toprotect worker health. The revisedprovision has been slightly modifiedfrom the provision in the NPRM that


read ‘‘* * * ensure proper air quality,quantity, and flow * * *’’ foratmosphere-supplying respirators. Theaddition of the words ‘‘* * * forbreathing air * * *’’ is to clarify thatunder no circumstances should air foratmosphere-supplying respirators be ofless than Grade D breathing air quality.

Paragraph (c)(1)(vii), as proposed,would have required employers toinclude ‘‘[t]raining of employees in therespiratory and health hazards of thehazardous chemicals to which they arepotentially exposed as required underthe Hazard Communication standard (29CFR 1910.1200).’’ Several commentersquestioned the need to cross-referencean existing OSHA standard in therespirator standard, and recommendedthat this provision be deleted (Exs. 54–154, 54–271, 54–278, 54–295, 54–307).OSHA agrees that the cross-reference isunnecessary, and the reference to theHazard Communication standard hasbeen removed from the final standard.However, the requirement thatemployers develop procedures thataddress the ‘‘Training of employees inthe respiratory hazards to which theyare potentially exposed during routineand emergency situations’’ remains,because there are respiratory hazards,such as biological hazards andradioactive particles, that are notcovered by the Hazard Communicationstandard.

Paragraph (c)(1)(viii) requiresemployers to develop procedures for thetraining of employees in the proper useof respirators, including putting on andremoving them, the limitations of thesedevices, and maintenance proceduresfor respirators. OSHA received nocomments on this provision, which hasbeen revised slightly since the proposalfor clarity.

Paragraph (c)(1)(ix) states that theprogram should include ‘‘Procedures forregularly evaluating the effectiveness ofthe program.’’ This provision isbasically the same as in the NPRMexcept that the word ‘‘periodically’’ hasbeen deleted to avoid the suggestionthat OSHA has a fixed interval in mind.This provision notifies employers thattheir written workplace proceduresmust include routine evaluation of theprogram to ensure that it is effective, up-to-date, and includes all necessaryprovisions. In workplaces whereworksite-specific conditions arerelatively stable, such as amanufacturing site, program evaluationmay be conducted on a fixed schedule.In other workplaces where worksiteconditions are less stable, employersmust develop schedules for evaluatingthe program that make sense in thatcontext.

In a general comment, the UnitedStates Enrichment Corporation (Ex. 54–283) stated that the final rule’srequirements for work procedures inparagraphs (c)(1)(i) through (c)(1)(ix)implied that OSHA intended separatedocuments to be developed to meet eachof the requirements, and asked OSHA toclarify this. It has always been OSHA’sintention that the employer can addressthe required program elements and thedevelopment of worksite-specificprocedures in a single document, thewritten respiratory protection program.OSHA believes that reorganizing theelements of this program to track theorder of the standard will facilitate theinclusion of all worksite-specificprocedures into one document.

In another general comment, PeterHernandez of the American Iron andSteel Institute (AISI) (Ex. 54–307) urgedOSHA to revise paragraph (c) and otherparagraphs of the final rule to removethe term ‘‘ensure,’’ which he interpretedas imposing an impossible burden onemployers. OSHA disagrees with thisinterpretation, however. OSHAstandards use the word ‘‘ensure’’because they impose a mandatoryrequirement to comply on employersand because the OSH Act andsubsequent case law have made it clearthat it is the employer’s responsibility tocompel compliance. The reasoningbehind this body of case law is that itis the employer, and not the employee,who controls the conditions of work ata given workplace. OSHA believes thatthe word ‘‘ensure’’ is appropriatebecause it indicates that the employermust manage, lead by example, train,direct, and, if necessary, set up adisciplinary system so that employeesunderstand that they must follow safeand healthful practices on the job.However, case law also makes it clearthat employers are not the ‘‘insurers’’ oftheir employees’ behavior. In otherwords, if an employer establishes,implements, trains employees in, andenforces safe operating procedures, anddoes so in a consistent manner, theemployer will not be liable for anemployee’s unforeseeable violation ofits safety rule.

Paragraph (c)(3) of the final rulerequires employers to designate aperson as program administrator and toensure that this person is qualified toperform the responsibilities of thisposition. The person can be qualifiedeither by appropriate training orexperience or both. The administrator isalso the person responsible forevaluating the program, as stated inparagraph (c)(3). This requirement isessentially unchanged from theproposal, although its language has been

clarified. The ANSI Z88.2–1992respiratory protection standard (Ex. 81)also contains a description of theresponsibilities of the programadministrator and a requirement that therespirator program be ‘‘periodicallyaudited to ensure that (a) the programprocedures reflect the requirements ofcurrent applicable regulations andindustry accepted standards and (b) theprogram as implemented reflects thewritten procedures’’ (See clause 5.3).The ANSI standard recommends thatthe audit be conducted by aknowledgeable person not directlyassociated with the program, rather thanby the program administrator. OSHAhas not adopted the ANSIrecommendation that periodic audits beperformed by knowledgeable outsidepersons because the OSHA standardrequires the administrator to bequalified to perform this task; thus, anadditional requirement for audits to beperformed by an outside party isunnecessary and may prove undulyburdensome for some employers.

The training requirements andexperience level necessary for theprogram administrator were the subjectof substantial comment. OSHAproposed that the program supervisor bea person ‘‘qualified by appropriatetraining and/or experience’’ to beresponsible for the respirator program.Many commenters supported thisperformance-based requirement (Exs.54–68, 54–80, 54–91, 54–175, 54–187,54–208, 54–219, 54–220, 54–222, 54–252, 54–319, 54–352, 54–361, 54–435,54–455). For example, the ServiceEmployees International Union (Ex. 54–455) supported the proposed‘‘performance-oriented qualifications forthe designated person (programadministrator).’’ Allied Signal (Ex. 54–175) stated that ‘‘there should be nospecific minimum training for programadministrators. We believe the level oftraining for the respirator programadministrator must be adequate to dealwith the complexity of the program.’’Motorola (Ex. 54–187) commented that‘‘Training requirements for thoseindividuals designated by the employerto administer the program should becommensurate with the type ofrespirator program needed at theworkplace.’’

Several commenters urged OSHA toadd a phrase to this requirement in thefinal rule to require that the level ofprogram supervisor training must beadequate to deal with the complexity ofthe program because the level oftraining appropriate for a workplacewith extensive respirator use issubstantially different from one withlimited respirator use (Exs. 54–175, 54–


187, 54–200, 54–206, 54–214, 54–219,54–222, 54–245, 54–265, 54–266, 54–275, 54–361). As Monsanto (Ex. 54–219)stated:

An employer’s respirator usage may belimited to dust respirators or may have awide variety of types covering both air-purifying and atmosphere-supplyingrespirators. Program administrator training/qualifications would need to cover a widerrange of topics in the latter case than in theformer case.

However, some commenters, e.g., theSparks Nevada Fire Department (Ex. 54–129), wanted to avoid imposing overlystringent requirements on choosing aprogram administrator, while others,e.g., the Grain Elevator and ProcessingSociety (Ex. 54–226), urged OSHA todelete the phrase ‘‘qualified by trainingand/or experience’’ on the grounds thatthere are no widely accepted criteria fordetermining such a programadministrator’s qualifications. A fewcommenters acknowledged that sincethe program administrator’s tasks oftenvary by type of workplace, it would bedifficult for OSHA to establish arequired minimum level of training thatwould be appropriate for all programsupervisors in all workplaces. MichaelRehfield, Safety Officer for theWestminster, Maryland Fire Department(Ex. 54–68) stated:

I am in total agreement that the personfulfilling this role and the ‘‘qualifications’’should be ‘‘performance oriented’’. Thatlanguage should appear in this section. It isimperative that the emergency responsecommunity be represented by performanceoriented standards or regulations since theassociated tasks are so diverse.

A working group from the StateUniversities of New York (Ex. 54–357)felt that the performance languageregarding program supervisors was toovague, and suggested that anonmandatory appendix be added toidentify the types of qualifications aprogram supervisor would need. TheUnited Automobile, Aerospace &Agricultural Implement Workers ofAmerica (UAW) (Ex. 54–387) wantedOSHA to define a body of knowledgenecessary to carry out the duties of aqualified program administrator.

OSHA discussed these qualificationsin the preamble to the NPRM at 59 FR58898–58899. That proposal discussionreiterated many of the points that aredescribed above: that the level oftraining appropriate for a workplacewith limited respirator use would bequite different from another withextensive use of different respiratortypes, and that the programadministrator can work with aworkplace respirator committee, orassign responsibility for portions of the

program to industrial hygienists, safetyprofessionals, or other respirator expertswhile retaining overall responsibility forthe program. In other words, the level oftraining of the program administratormust be adequate to deal with thecomplexity of the respirator program.

The AFL–CIO (Exs. 54–428, 255)urged OSHA to add a new definition toparagraph (b) for qualified person asfollows:

Qualified Person: This should be definedas, someone who is capable of identifyingexisting and predictable respiratory hazardsin the workplace and who maintains acommon knowledge of the respiratorstandard. This individual should possess theauthority to take prompt corrective action toeliminate hazards including the measuresrequired in subsection (c). The qualifiedperson shall be certified by themanufacturer(s) for their ability to select andmaintain the type(s) of respirator(s) that is/are used on the job site or possess theexperience and knowledge needed toproperly select respirators for the employeesand job situation.

Instead of adopting the AFL–CIOdefinition for ‘‘qualified person,’’ OSHAhas relied on the type of wording usedin the ANSI standard, which is moreperformance oriented. Specifying indetail the type and extent of trainingrequired for program administratorsdepends upon the type of workplaceand is best left to the employer, inOSHA’s opinion. For example, the levelof training that would be appropriate fora workplace with limited respirator usewould be quite different from thatrequired at another workplace withextensive respirator use for IDLHatmospheres, highly toxic chemicals, orother complex respirator use operations.Therefore, OSHA has adopted adefinition of training and experiencethat uses performance language and issimilar to the ANSI Z88.2–1992standard’s requirement. However,OSHA does require employers to ensurethat the level of training for therespirator program administrator isadequate to deal with the complexity ofthe workplace.

In keeping with this approach, OSHAhas not established any one trainingprogram, such as the NIOSH respiratorcourse, as the level of training programadministrators must achieve. OSHAbelieves that NIOSH’s course isexcellent, and therefore more thansufficient in most cases. However,OSHA acknowledges commenters’concerns that a general respiratortraining course covers a broad range ofmany different respirator types anduses, and provides information that isnot tailored to any one particularworkplace (Exs. 54–220, 54–265, 54–

342, 54–435). Typical of thesecomments is one by the United ParcelService (Ex. 54–220), which stated: ‘‘Anattempt to fashion uniform standards forall administrators of all respiratoryprograms could result in inadequatetraining for administrators ofparticularly sophisticated or specializedprograms and irrelevant training foradministrators of relatively simpleprograms.’’ The North AmericanInsulation Manufacturers Associationagreed, stating (Ex. 54–342) ‘‘Arequirement that supervisors undergo arigid minimum training regimen, whichwould require instruction on manyissues irrelevant to the supervisor’s ownsituation, would be excessive andbeyond the rule’s intended objective.’’For example, extensive training oncertain types of respirators such asSCBAs would be inappropriate forprogram administrators with simpleprograms that don’t use SCBAs. In othercases, respirator program administratorswith highly complex respiratorprograms may need an even morecomprehensive course than thatprovided by a general respirator trainingcourse. Based on the above discussion,OSHA has retained a performance-basedprogram approach. OSHA anticipatesthat larger establishments will developtraining requirements for respiratorprogram administrators that fit theneeds of a workplace-specific respiratorprogram.

OSHA has prepared a Small EntityCompliance Guide setting forth how asmall business owner, manager or anemployee of the small business can bequalified to be a program administrator.It also sets forth a sample respiratorprogram to guide small businesses. Ifthe employees of a small business areonly exposed to nuisance dusts andrelatively non-toxic chemicals and useonly a few types of relatively simplerespirators, knowledge of the guide andmaterials supplied by the respiratormanufacturer may be sufficient for thesmall business owner or an employee tobecome qualified as a programadministrator. If more dangerouschemicals or high exposures are present,or sophisticated respirators are used, theprogram administrator must have moreknowledge or experience. In thesecircumstances, it may be necessary forthe administrator to seek out theexpertise needed or to obtainappropriate training.

The need for a specific individual tobe in charge of the respirator programwas discussed by several commenters.One commenter argued that requiringthat a specific person be selected asprogram administrator requires theequivalent of a full-time person to


manage the program and conductperiodic reviews of its performance (Ex.54–160). Motorola (Ex. 54–187) statedthat one overall program administratorwould be a problem for decentralizedworkplaces. Motorola recommendedthat OSHA permit a committee ormultiple employees to be responsiblefor the respirator program, thus allowingthe employer to tailor the program tomeet the needs of each particularworkplace. Dow (Ex. 54–278) alsosupported the use of a committee orteam with joint responsibility for therespirator program at large sites. DukePower (Ex. 54–326) stated that at largefacilities, such as nuclear stations, it isoften necessary to designate more thanone program administrator to addressradiological and non-radiological use ofrespirators. The Public Service Electricand Gas Company (Ex. 54–196) said itmay be more effective to have a programadministrator for each ‘‘business unit’’in a decentralized, diversified company,particularly where each unit’srespiratory protection needs aredifferent (Ex. 54–196). The AFL–CIO(Ex. 54–428) wanted to have onequalified person responsible for theprogram, with a ‘‘site person’’ at eachwork site, who would be responsible forthe program at that site, but who wouldreport to the qualified person. TheDepartment of Defense (Ex. 54–443),specifically the Navy, urged OSHA toadd language to require that each‘‘activity’’ designate a personresponsible for the respiratoryprotection program because a singleprogram administrator would be apotential problem for a large, multi-tiered employer with activitiesthroughout the world, such as the Navy.

The final standard continues torequire that a person qualified bytraining or experience be designated tobe responsible for the overallmanagement and administration of theprogram to ensure that the integrity ofthe respiratory protection program ismaintained through the continuousoversight of one responsible individual.The program administrator may servelargely in an oversight and coordinationrole between the various subunits ordepartments that perform duties insupport of the respiratory program.Regardless of the number of subunits,each employer must ensure that allsubunits report to one overall programadministrator for coordination of theprogram. The program administrator canuse the assistance of industrialhygienists, safety professionals, or otherrespirator experts to help run therespirator program. The programadministrator can work with a

committee or assign responsibility forportions of the program to otherpersonnel, but the overall responsibilityfor the operation of the program mustremain with the designated programadministrator. This approach promotescoordination of all facets of the program.For large companies or multipleworksites, the program administratorcan delegate to a qualified person theresponsibility for the day-to-dayoperation of the program at a specificsite or for a specific activity. However,coordination between differentworksites is an important aspect of theoperation of a good program; therefore,ensuring implementation of the overallrespirator program remains the duty andresponsibility of the programadministrator. For small and moderatesized employers, OSHA believes thatthe duties of a program administratorwill require only a small part of oneemployee’s time.

Paragraph (c)(4) of the final rulerequires employers to providerespirators at no cost to the employee.This was included in the proposal inparagraph (d)(1) and has been moved toparagraph (c) of this final standard. Thisprovision reflects OSHA’s strongorientation that the costs of complyingwith safety and health requirementsmust be borne by the employer. OSHAhas a long-standing policy thatemployers are obligated to provide andpay for necessary personal protectiveequipment (PPE) such as respiratorsused by employees on the job. Acompliance memorandum of October18, 1994, titled ‘‘Employer Obligation toPay for Personal Protective Equipment’’provides detailed guidance on thisissue. It is available online on theInternet on OSHA’s home page athttp://www.OSHA.gov. The inclusion ofthis provision is consistent with recentOSHA standards, e.g., Cadmium, 29CFR § 1910.1027; 1,3-Butadiene, 29 CFR1910.1051; and Methylene Chloride, 29CFR 1910.1052.

OSHA is aware that the OccupationalSafety and Health Review Commissionhas not always agreed with the Agencythat standards requiring an employer to‘‘provide’’ safety or health equipmentalso require the employer to pay for thatequipment. See, e.g., Union Tank CarCo., OSHRC No. 96–0563 (October 16,1997). OSHA believes the Commissionis wrong about this issue. OSHA intendsthe language ‘‘at no cost to theemployee’’ in paragraph (c)(4) to makethe employer’s obligation to pay for therespiratory protection required by thisstandard crystal clear.

The requirement that the employerbear the costs of employee training andmedical evaluations has also been

moved to paragraph (c)(4) of the finalrule, in order to consolidate all similarprovisions of the standard that clarifythat, for these provisions, there is nocost to the employee. Section 6(b)(7) ofthe OSH Act requires that employersprovide medical exams and evaluationsat no cost to employees.

Paragraph (d)—Selection of Respirators

OverviewParagraph (d) of the final rule

contains respirator selection criteria andrequirements. OSHA has included theseprovisions in the final rule because therecord contains many examples ofworkers using respirators that areinappropriate for the type of respiratoryhazards present (e.g., wearing paperdust masks where the exposure is to agas or vapor contaminant (UAW, Ex.54–387); using half facepiece respiratorsin acrylonitrile IDLH atmospheres of 20ppm (International Chemical WorkersUnion (ICWU), Ex. 54–427)). Inaddition, OSHA’s long enforcementexperience has shown that employersoften lack the information necessary tomake informed choices about respiratorselection. OSHA stated in the proposal(59 FR 58899) that a major deficiency ofthe previous standard is that it did notcontain selection criteria; instead, itmerely referred employers to the ANSIZ88.2–1969 standard.

No participant in this rulemakingdisagreed with OSHA’s decision that thefinal standard should includemandatory selection criteria. The recorddoes show, however, that there aredifferences of opinion about howrestrictive and comprehensive therequired criteria should be, and howmuch flexibility should be left toemployers in the selection process. Forexample, the Association of AmericanRailroads (Ex. 54–286) stated that thedetails of respirator selection should beleft to the regulated community and thatOSHA should only specify the outcomedesired, while the Service EmployeesInternational Union (SEIU) (Ex. 54–455)commented that OSHA should‘‘strengthen the wording to make it clearemployers must obtain and account forall of the factors listed.’’ OSHA believesthat those employers who employ on-site occupational health professionalsgenerally have the expertise to selectrespirators that are appropriate for theirworkers. The record contains a numberof examples of well-thought-outselection programs (e.g., Exs. 142, 155,163). These examples show that thecurrent practice of many employersalready conforms to the selectionrequirements of paragraph (d). For otheremployers, however, clearly stated


respirator selection rules and guidanceare required.

OSHA notes that advice on theselection of respirators is available frommany sources. NIOSH has developed arespirator decision logic, widelyavailable and used since 1987, whichprovides a schematic selection guidecovering all critical areas of respiratorselection (Ex. 9). The selection guide forthe ANSI Z88.2–1969 respiratorstandard was incorporated by referenceinto the previous OSHA standard, andthe 1992 Z88.2 ANSI standard containsupdated and comprehensiverecommendations on respiratorselection. OSHA believes thatemployers will find useful informationin each of these guides on varioustechnical problems that this standardmay not cover explicitly. In addition,information is provided by respiratormanufacturers who publish selectionguides relating to their models (See, e.g.,Mine Safety Appliances Company(MSA) Respirator Selection Guide, Ex.150; and ISEA’s Respirator BuyersGuide and Safety Video Resource List,referenced in Ex. 147). Manufacturersalso provide selection advice throughtelephone help lines, sales staff, verbalcommunications or distribution ofcompany product information, and on-site evaluations of product use (See, e.g.,Tr. at 1438–1439). Chemicalmanufacturers also provide informationabout respirator selection to help thepurchasers of their products (See CMA,Tr. 1726–7; Union Carbide Corporation,Ex. 54–255).

Because of the variety and detail ofselection information available, OSHAbelieves it is necessary in the final ruleto specify broad performance criteria, inaddition to a few specific rules relatingto highly hazardous operations (i.e.,IDLH situations). The final rule setsforth general rules for selectingrespirators for routine operations,prescribes specific kinds of respiratorsfor identified highly hazardousatmospheres and emergency situations,and specifies when air-purifyingrespirators can reliably be used. OSHAchose not to specify in the regulatorytext all the situations and respirator-related factors that an employer shouldconsider but instead to stateperformance objectives. Only forworkplace situations widely accepted ashighly hazardous, such as thoseassociated with IDLH atmospheres, doesthe standard require maximallyprotective respirators.

Because paragraph (d) does notaddress in detail all the relevant factorsthat may affect employers’ selection ofparticular respirators, employers shouldrely on other information sources to

ensure that the respirators they selectare appropriate for conditions in theirspecific workplaces. Respiratormanufacturers are the source of muchuseful information, and the record ofthis rulemaking indicates that much ofthis information is both helpful andreliable. Indeed, market mechanismswork to encourage the dissemination ofaccurate information. OSHA expectsthat smaller employers will thusgenerally be able to rely on the technicalassistance provided by manufacturerson respirator selection and that doing sowill mean that they will usually be incompliance with this standard. Forthese reasons, paragraph (d)concentrates on the minimum selectioncriteria that the record shows must beadhered to by all employers whenselecting respirators for their employees’use.

In the following provision-by-provision summary and explanation,OSHA explains the changes reflected inthe final rule, both from the provisionsproposed and those in the Agency’sprevious respiratory protection standard(§ 1910.134).

Paragraph (d)(1)—General RequirementsParagraph (d)(1) prescribes general

rules that apply to the selection of allrespirators. Paragraph (d)(1)(i) requiresthe employer to select and provide anappropriate respirator based on therespiratory hazard(s) to which theworker is or will be exposed and on theworkplace and user factors that have thepotential to affect respiratorperformance and reliability. Thisprovision continues a requirement fromthe previous standard: (‘‘respiratorsshall be selected on the basis of hazardsto which the worker is exposed’’(§ 1910.134(b)(2)) and clarifies that thehazard must be viewed in the context ofthe workplace and worker conditionsthat may reduce or impair theeffectiveness of a respirator otherwiseappropriate for the hazard. There isgeneral agreement that taking workingconditions into account is crucial toproper respirator selection: a respiratorthat is protective under some conditionsof wear will fail under others, while arespirator that is appropriate for a givenhazard may not be workable in aparticular workplace (e.g., an airsupplied respirator in a tightlyconfigured space). For example, aworker wearing SCBA who is requiredto perform extremely heavy work maydeplete the air supply of the respiratorwell before its calculated service life isreached. This means that the employermust evaluate the employee’s level ofexertion in order to determine whetherto choose a supplied-air respirator

rather than a SCBA. The recent ANSIstandard also states that the purpose ofrespirator selection is to determinewhich respirator type or class will offer‘‘adequate protection’’ (ANSI Z88.2–1992).

Final paragraph (d)(1)(i) also requiresemployers to consider workplace anduser factors that may affect therespirator’s performance and reliabilitywhen making a respirator selection.Although other paragraphs of thestandard address the major factorsaffecting respirator performance, i.e., fit,faceseal leakage, and maintenance andcleaning, factors specific to the job, user,or worksite often play an important rolein respirator performance. OSHA notedin the proposal (59 FR 58900) that workactivities and factors such astemperature and humidity ‘‘also affectthe stress level associated with wearinga respirator as well as the effectivenessof respirator filters and cartridges;employees using respirators for longerperiods of time [under such stressfulconditions] may need different types ofrespirators for more comfortable wear.’’

Similarly, where the respirator-wearing employee must communicatewith other workers, perhaps to warnthem about the presence of workplacehazards, the respirator must allow theemployee to perform this vital function.OSHA thus agrees with ANSI that ‘‘it isimportant to ensure that respiratorwearers can comfortably communicatewhen necessary, because a worker whois speaking very loudly or yelling maycause a facepiece seal leak, and theworker may be tempted to temporarilydislodge the device to communicate’’(ANSI Z88.2–1992, clause A.13).Therefore, for example, the employermust ensure that speaking will notinterfere with the fit of the negative-pressure elastomeric respirator selected.If the employees are using PAPRs orSCBA, amplification devices, includingspeaking diaphragms and microphones,that can be worn with the respirators areavailable.

The proposal (59 FR 58900) notedanother example in the proposal ofworksite conditions that could affectrespirator selection: ‘‘* * * airlinerespirators should not be used bymobile employees around movingmachinery unless entanglement ofairlines in equipment is easily avoided.’’Employers have always been requiredby OSHA to consider such factors asthese, because paragraph (a)(2) of theprevious respirator standard requiredemployers to select respirators that are‘‘applicable and suitable for the purposeintended.’’

Paragraph (d)(1)(i) applies wheneveremployers provide respirators to their


employees and require their use,whether or not an OSHA standardmandates respirator use in the particularenvironment. The preamble discussionrelating to paragraph (c)(1) discussesemployer-required respirator use inmore detail and explains OSHA’sreasons for reaching this conclusion.

Paragraph (d)(1)(ii) requires theemployer to select a NIOSH-certifiedrespirator and to use the respirator onlyin ways that comply with the conditionsof its certification. There was littlecontroversy about this requirement, andthere is no disagreement that respiratorsmust be tested and found to be effectivebefore they can be marketed. NIOSH hasperformed this function in the past andhas begun to revise its certificationrequirements to ensure that itsprocedures continue to define theperformance capabilities of acceptablerespirator models, and to identifyunacceptable models. The ISEA (Ex. 65–363), the trade association thatrepresents most major respiratormanufacturers, urged OSHA to requirethat only NIOSH-certified respirators beused to comply with this standard, andother commenters agreed (Exs. 54–187,54–213, 54–387, 54–428).

The wording of this provision of thefinal rule differs slightly from that of theproposed provision. The proposalwould have required that only NIOSH‘‘approved and certified’’ respirators beselected. For clarity, the reference toNIOSH-approved respirators has beenreplaced in the final rule by arequirement that respirators be usedonly in accordance with the conditionsof their certification. NIOSH approvesrespirators by certifying them; however,some certifications contain conditionslimiting the situations in which therespirator may be used. This issometimes described as NIOSH‘‘approval’’ of the respirator for aparticular use.

Increasingly, however, NIOSH doesnot certify respirators for specific uses.For example, NIOSH does not currentlycertify respirators for use againstbiological hazards. Where NIOSH hasnot specifically certified any respiratorfor use against the particularcontaminant present in the workplace,the employer must select a NIOSH-certified respirator that has nolimitation prohibiting its use againstthat contaminant. The respirator mustbe appropriate for the contaminant’sphysical form and chemical state andthe conditions under which it will beused. All respirators must be chosenand used according to the limitations ofthe NIOSH certification, which appearson the NIOSH certification label.

The requirement for NIOSHcertification is unconditional in thefinal standard, as it was in the proposal.However, because OSHA stated in theproposed preamble that thisrequirement would apply only whensuch respirators ‘‘exist’’ (59 FR 58901),some commenters urged OSHA to statein the regulatory text that therequirement for NIOSH certificationapplied only to existing certifications(See, e.g., Ex. 54–434). For example, theDepartment of the Army (Ex. 54–443)urged OSHA to permit the use ofrespirators not approved by NIOSH insituations where another authority hasjurisdiction and the documentation toattest to the adequacy of the respirator’seffectiveness against the contaminant ofconcern. The Army (Ex. 54–443D) statedthat its employees and contractors maybe exposed to certain ‘‘military uniquecontaminants’’ for which no NIOSH-approved respirator exists but for whichmilitary respirators, e.g., gas masks,have specifically been developed andtested and are being used by civilianand contractor personnel in operationssubject to OSHA’s jurisdiction. TheArmy urged OSHA to include in thestandard ‘‘approval authority of theSecretary of the Army for militaryrespirators * * * for which no NIOSHapproved respirator exists’’ (Ex. 54–443D).

OSHA recognizes that there areunique contaminant situations, such asthose involving chemical warfareagents, that involve primarily militaryexposure and that may requirespecialized respiratory protectionequipment. NIOSH certification forrespiratory protection specific to suchhazards does not exist and is not likelyto be forthcoming. OSHA also notes,however, that, although the Departmentof the Army argued strongly for OSHArecognition of Army authority to testand approve respirators, the Departmentof the Air Force commented that it usesonly NIOSH-certified respirators, andrequested no exception (Ex. 54–443A).OSHA will examine on a case-by-casebasis those situations involving civiliancontractors whose employees wear non-NIOSH tested respirators that theybelieve protect employees adequatelyand that have been tested and approvedby other Federal agencies for use againstunique contaminants.

A similar comment was raised byDOE regarding radioactive hazards (Ex.54–215). DOE stated that, in the nuclearindustry, no NIOSH-certified respiratorexists for tritium applications andworkers therefore must wear non-approved supplied-air suits; thisequipment has been tested by LosAlamos National Laboratory, and the

suits have been successfully used formany years. The DOE administers itsown job-by-job approval system forthese suits. OSHA’s authority to enforcethe Agency’s safety and healthstandards at gaseous diffusion plantsowned by DOE and leased to the UnitedStates Enrichment Corporation wasestablished legislatively in 1992, andOSHA has recently completed amemorandum of understanding withDOE on this issue (60 FR 9949, Jan. 31,1995). OSHA is currently evaluating anapplication from one of these facilitiesfor a variance relating to these suits. Thecriteria set out in Section 6(d) of theOSH Act will govern this determination.OSHA is not determining theacceptability of supplied-air suits aspart of this rulemaking proceeding,because the Agency believes thevariance proceeding, which can focuscloser attention on the strengths andlimitations of these suits for theparticular use situations, is theappropriate forum to decide this issue.

OSHA notes that NIOSH certificationis a minimum qualification. Theemployer must still assess whether therespirator meets all other selectioncriteria in this standard before it can bechosen for a particular application. Forexample, as pointed out by an exchangewith Richard Duffy of the InternationalAssociation of Fire Fighters (IAFF),NIOSH representatives acknowledgedthat the employer must evaluatewhether NIOSH-certified equipmentwill withstand the specificenvironmental conditions forfirefighting because NIOSH flow raterequirements do not consider thestresses involved in firefighting, nordoes NIOSH currently evaluaterespirators for their ability to withstandthose stresses (Tr. 364–365).

In his testimony at the OSHAhearings, Richard Duffy of the IAFFrecommended that OSHA require thatSCBAs used in firefighting meet therequirements of the National FireProtection Association’s NFPA–1981Standard on Open Circuit BreathingApparatus (Tr. 455). This NFPAstandard establishes more stringentperformance criteria for SCBAs used infirefighting than those currently used byNIOSH. NIOSH recognizes that itscurrent 42 CFR 84 respiratorcertification standards may not beprotective enough for respirators used infirefighting. In an October 7, 1997 letterto all manufacturers and interestedparties, NIOSH announced its intent todevelop new technical modules toupdate 42 CFR 84. One of the proposedtechnical modules to which NIOSHintends to give priority treatment willaddress SCBAs, including the


incorporation of NFPA performancerequirements for SCBAs. NIOSH alsointends to propose an Administrative/Quality Assurance module on the use ofindependent testing laboratories in thecertification program, another issueraised by commenters in thisproceeding. OSHA believes that NIOSHwill resolve any deficiencies in itscurrent respirator certification standardsthrough these new 42 CFR 84rulemaking modules. OSHA simply isnot equipped to take on the respiratorapproval and certification processcurrently performed by NIOSH.Therefore, the final OSHA respiratorstandard continues to require the use ofNIOSH-certified respirators and doesnot incorporate the NFPA performancerequirements for SCBAs.

OSHA believes that carving out evenlimited exceptions to NIOSH control ofrespirator certification authority wouldconfuse the regulated community andwould not resolve the needs of the vastmajority of respirator users. Commentsby respirator users and workerrepresentatives support OSHA’s finaldecision (See, e.g., Exs. 54–265, 54–118,54–213, 54–387, 54–455). The final rule,in paragraph (h), also requires that whenrespirator parts are replaced or changed,the replacement parts must be NIOSHcertified.

In the proposal (59 FR 58901), OSHAstated that developing an OSHArespirator approval mechanism to fill inthe gaps in NIOSH certification wouldnot be an efficient use of governmentresources. Nonetheless, the Agencyasked for comment on this issue. Therewas no consensus among theparticipants who commented on thispoint. Some commenters supported anOSHA role in approval on a temporarybasis, while an employer waits forNIOSH approval, or an alternativegovernmental approval process (Exs.54–213, 54–346, 54–443). Still othersopposed OSHA’s involvement in anapproval process (Exs. 54–278, 54–265,54–118, 54–213, 54–387, 54–455). Thefinal rule is therefore similar to theproposal, which also discussed limitedalternatives to NIOSH certification andconcluded that ‘‘it is inappropriate forOSHA to try to correct problems withpresent NIOSH/MSHA regulations inthe revised respirator standard’’ (59 FR58891).

OSHA believes that NIOSH hasfocused on closing any gaps in itscertification program. NIOSH’s abilityand experience in this area areunparalleled, and OSHA believes thatNIOSH can best resolve any concernsthrough its own proceedings. Further, asstated in the proposal, OSHA lacks theresources to perform respirator testing.

OSHA will, however, continue toevaluate, on a case-by-case basis,whether variance or complianceinterpretations are appropriate in caseswhere employers claim that there are noNIOSH-certified respirators for use in aparticular situation.

Paragraph (d)(1)(iii) of the final rulerequires the employer to identify andevaluate the respiratory hazard(s) in theworkplace. To perform this evaluation,the employer must make a ‘‘reasonableestimate’’ of the employee exposuresanticipated to occur as a result of thosehazards, including those likely to beencountered in reasonably foreseeableemergency situations, and must alsoidentify the physical state and chemicalform of such contaminant(s). Whereconditions are such that the employercannot carry out such an evaluation,e.g., where exposure monitoring or othermeans of estimation cannot be used,paragraph (d)(1)(iii) requires theemployer to treat the atmosphere asIDLH. Many of the components ofparagraph (d)(1)(iii) of the final standardhave been required practice since 1971because they were included in theselection provisions of the 1969 ANSIstandard incorporated by reference intoOSHA’s previous respiratory protectionstandard. Paragraph (d)(1)(iii) of thenew standard makes these provisionsclearer by stating them explicitly in theregulatory text.

Identifying and evaluating the hazardsa respirator is to provide protectionagainst clearly play a pivotal role inrespirator selection. For example,according to ANSI, ‘‘Respirator selectioninvolves reviewing each operation to* * * determine what hazards may bepresent (hazard determination)’’ (ANSIZ88.2–1992, clause 7.2.2; See also AISI,Tr. 639). Many other commentersemphasized the important role of hazardidentification in respirator selection(Exs. 54–168, 54–181, 54–186, 54–208,54–234, 54–273, 54–307, 54–327, 54–346, 54–426, 54–428). Once anemployer identifies the nature of therespiratory hazard or hazards present,the employer must evaluate themagnitude of the hazard to determinethe potential exposure of each employeeand the extent to which respirators ofvarious types can reduce the harmcaused by that exposure.

There was extensive comment on theselection process outlined in theproposed paragraph dealing with hazardevaluation (Exs. 54–154, 54–168, 54–181, 54–202, 54–219, 54–245, 54–278,54–428). Commenters representingworkers generally supported thedetailed approach taken in the proposaltoward hazard evaluation. For example,the Service Employees International

Union ‘‘support[ed] the detailed list offactors to be considered in respiratorselection * * * [which] successfullyincorporates the important frameworkfrom the NIOSH decision logic criteriain an easy-to-understand form’’ (Ex. 54–428).

Some commenters, however (Exs. 54–154, 54–168, 54–181, 54–219, 54–245,54–278), stated that the scope and depthof the hazard evaluation and the itemsto be covered should be left to thediscretion of the employer. For example,the Eastman Chemical Company (Ex.54–245) and the Dow ChemicalCompany (Ex. 54–278) requested thatOSHA make the requirement‘‘performance oriented’’ and ‘‘flexible’’;the Department of the Navy, PortsmouthNaval Shipyard (Ex. 54–154), noted thatdetailed analysis for each work situationis not necessary for shipbuilding, andthat the timing and content of anappropriate evaluation vary.

In response to these comments, OSHAhas revised paragraph (d)(1)(iii) to bemore performance oriented; thisprovision of the final standard no longerspecifies precisely how employers are toconduct the required evaluation. Theproposal (at paragraph (d)(3)) wouldhave required employers to ‘‘obtain andevaluate’’ information on eleven specificfactors for each work situation. Theseproposed factors were the nature of thehazard; its physical and chemicalproperties; its adverse health effects; theoccupational exposure level; the resultsof workplace sampling; the workoperation; the time period of respiratorwear; the work activities and stresses onthe wearer; fit test results; warningproperties; and the capabilities andlimitations of respirator types. AlthoughOSHA continues to believe that each ofthese factors is relevant to respiratorselection under some circumstances, areview of the record has convincedOSHA that each factor is not crucial inevery respirator selection process andthat the proposed requirement wouldhave led to needless duplication ofeffort and unnecessarily detailedevaluations.

The Oil, Chemical and AtomicWorkers International Union (OCAW)(Ex. 54–202) urged OSHA to require awritten hazard assessment each timethat a respirator was selected. Paragraph(d)(1)(iii) of the final rule does notrequire a written assessment; this wasnot proposed, and OSHA believes thatemployers should be free to adopt thebest approach for justifying theirrespirator selections, based on thehazard assessment. The final rulerequires the employer to identify andevaluate the respiratory hazards present,determine their physical state and


chemical form (e.g., whether they arepresent in the form of a gas or vapor;what their valence state or condition is,where relevant), and assess themagnitude of the hazard they present toworkers under normal conditions of useand in reasonably foreseeableemergency conditions.

OSHA finds that it is essential foremployers to characterize the natureand magnitude of employee exposuresto respiratory hazards before selectingrespiratory protection equipment. Thelanguage contained in paragraph(d)(1)(iii) of the final rule does notspecify how the employer is to makereasonable estimates of employeeexposures for the purposes of selectingrespirators, nor does the standardrequire the employer to measure workerexposures to airborne hazards. OSHAhas always considered personalexposure monitoring the ‘‘goldstandard’’ for determining employeeexposures because this is the mostreliable approach for assessing howmuch and what type of respiratoryprotection is required in a givencircumstance. This general view is alsoshared by the industrial hygienecommunity. All of OSHA’scomprehensive substance-specifichealth standards have requiredemployee exposure monitoring todetermine both the effectiveness ofexisting control measures and the typeof respiratory protection needed.

OSHA continues to hold this viewwith regard to assessing employeeexposure in connection with thisrespiratory protection standard.However, OSHA recognizes that thereare many instances in which it may notbe possible or necessary to take personalexposure measurements to determinewhether respiratory protection isneeded. Although sampling andanalytical methods exist for the vastmajority of substances for which OSHAhas a PEL (29 CFR 1910.1000), there arenumerous other substances for whichthere are no readily available methodsfor personal sampling. In other cases,the nature of the materials and productsbeing used in the workplace, and theway in which they are used, make ithighly unlikely that an employeeworking with them would be exposed ina manner that would make respiratoryprotection necessary. In these kinds ofsituations, the final rule permitsemployers to use other approaches forestimating worker exposures torespiratory hazards.

For example, employers may rely oninformation and data that indicate thatuse or handling of a product or materialcannot, under worst-case conditions,release concentrations of a respiratory

hazard above a level that would triggerthe need for respirator use or require useof a more protective respirator. Thisapproach is similar to that used inseveral OSHA substance-specific healthstandards, which permit employers touse objective data in lieu of exposuremonitoring to demonstrate that theiremployees cannot be exposed above anaction level (See, for example, 29 CFR1910.1027, Cadmium; 1910.1048,Formaldehyde; 1910.1047, EthyleneOxide; 1910.1028, Benzene). Objectivedata can be obtained from an industrystudy or from laboratory test resultsconducted by manufacturers of productsor materials being used in theworkplace. To generalize from data inan industry-wide survey to conditionsin a specific workplace, the survey musthave obtained data under conditionsclosely resembling the processes, typesof materials, control methods, workpractices, and environmental conditionsin the workplace to which it will begeneralized, i.e., the employer’soperation.

Data from industry-wide surveys bytrade associations for use by theirmembers, as well as from stewardshipprograms operated by manufacturers fortheir customers, are often useful inassisting employers, particularly small-business owners, to obtain informationon employee exposures in theirworkplaces. For example,representatives of the North AmericanInsulation Manufacturer’s Association(NAIMA) testified (Tr. 597) that * * *‘‘[w]e have conducted numeroussurveys on end use customers,conducted research with Johns HopkinsUniversity, for example to provideestimates of routine exposures and* * * those data, when collectedappropriately and with organized laborand with other industry groups, * * *can assure that the right respirator isselected.’’ NAIMA stated (Tr. 616, 618),‘‘it is ultimately the employer’sresponsibility’’ to evaluate whether dataprovided by suppliers or others relate totheir workplace conditions andoperations. However, it is clear thatsuch programs can often assistemployers to estimate workplaceexposures reliably enough to makecorrect respirator choices without theneed for employee monitoring.

Another approach that can be used byemployers to estimate employeeexposures involves using mathematicalapproaches and obtainable information.Employers can use data on the physicaland chemical properties of aircontaminants, combined withinformation on room dimensions, airexchange rates, contaminant releaserates, and other pertinent data,

including exposure patterns and workpractices, to estimate the maximumexposure that could be anticipated inthe workplace. Methods that utilize thisapproach are readily available in severaltextbook sources; for example, theACGIH Industrial Ventilation Manualcontains calculations that can beapplied to certain situations to estimateworker exposures. Relying on such anapproach to estimate exposures requiresthe use of safety factors to account foruneven dispersion of the contaminant inthe air and the proximity of the workerto the emission source. Usually, thisapproach works best in situations whereemployees use small amounts of achemical product intermittently, orwhere contaminant releases are fairlyconstant and predictable. This approachmust be used continuously, and the dataobtained should therefore be interpretedconservatively (i.e., should err on theside of worker protection).

In workplaces involving manycomplex factors, the use of estimationtechniques to characterize workerexposure is associated with a highdegree of uncertainty. In theseinstances, OSHA recommends thatemployers conduct exposure monitoringinstead of relying on estimationtechniques because they will then beable to have confidence that theappropriate respiratory protectiondevice has been selected and that theyare in compliance with the standard.Furthermore, OSHA believes that inworkplaces where many complexfactors add uncertainty to exposureestimates obtained through modeling,employers will find it easier and lesscostly to conduct personal exposuremonitoring to evaluate the need forrespiratory protection.

Many commenters urged OSHA not tospecifically require monitoring in thestandard because other means ofassessing potential exposures areavailable (Exs. 54–153, 54–208, 54–219,54–237, 54–273, 54–307, 54–327, 54–443). These participants asked theAgency instead to adopt the approachtaken in the ANSI standard Z88.2–1992,clause 7.2.2.1(e), which allowsemployers to estimate, as well asmeasure, exposures in the workplace.One commenter questioned the utility ofexposure monitoring data for respiratorselection because exposure samplingprovides only a ‘‘snapshot’’ of hazardson any given day (Ex. 54–178). Othercommenters disagreed, however. Forexample, Scott Schneider (Tr. 1520) ofthe AFL–CIO stated, ‘‘In mostworkplaces that I’ve been in there reallyis very, very little exposure data toknow how much a person is exposed to* * * exposures are quite variable from


day to day. And from worker toworker.’’ (See comments to same effectby OCAW, Ex. 54–202.) Someparticipants specifically asked OSHA tomake workplace sampling of airborneconcentrations of contaminants explicit(Tr. 1009 and Ex. 54–428; Ex. 54–427).

That some exposure monitoringresults may be inadequate begs thequestion of whether adequatemonitoring should be conducted.OSHA’s experience in enforcingpermissible exposure limits in the AirContaminant standard, 29 CFR1910.1000, and for substance-specificstandards, confirms that, unlessoperations are highly repetitive,conditions are constant, and estimatesbased on ‘‘historical’’ and ‘‘objectivedata’’ are made by experiencedindustrial hygiene professionals, mostemployers need exposure monitoringresults to estimate employee exposurelevels reliably. OSHA enforcementexperience also demonstrates that,where exposures are highly variable,fragmentary monitoring results maymislead employees and employers,unless they are based on competentsampling strategies. The frequency andduration of monitoring, therepresentativeness of the employees andoperations sampled, and the skill withwhich sampling and analysis areperformed all influence the reliability ofmonitoring results. In makingreasonable estimates of employeeexposures to satisfy the requirementscontained in paragraph (d)(1)(iii), OSHAexpects employers to account forpotential variation in exposure and torely on data or information that reflectsuch variation. This is accomplished byusing exposure data collected with astrategy that recognizes exposurevariability, or by using worst-caseassumptions and estimation techniquesto evaluate the highest foreseeable levelsto which employees may be exposed.The hazard assessment requirements infinal paragraph (d)(1)(iii) carry overfrom the requirement of the previousstandard, which incorporates byreference the ANSI Z88.2–1969 (clause6.2) statement that ‘‘[a]ny erring in theselection of respirators shall be on thesafe side.’’

Paragraph (d)(1)(iii) also requires anemployer to consider the environmentIDLH if employee exposures cannot beestimated reasonably. This provision isintended to address those limitedsituations where neither exposuremonitoring, professional judgment, norestimation techniques can be relied onto reliably select adequate respiratoryprotection equipment. This provisionreflects a similar one in the 1992 ANSIstandard, which requires atmospheres to

be considered IDLH if it is not possible‘‘to determine what potentiallyhazardous contaminants may be present* * * or if no exposure limit orguideline is available, and estimates oftoxicity cannot be made’’ (ANSI Z88.2–1992, clause 7.2.2.2 (b)(c)).

Several commenters (Exs. 54–381, 54–352, 54–267) objected to OSHA’sproposed requirement that atmospheresbe considered IDLH ‘‘where theconcentration of the hazardous chemicalis unknown’’ (59 FR 58939), and statedthat it would be neither practical nornecessary to wear positive pressurerespirators in all such situations (Ex.54–352). One commenter believed thatrequiring the most protective respiratorsfor ‘‘every unknown hazardouschemical atmosphere’’ would result in95 percent of the workforce beingrequired to use them (Ex. 54–267).OSHA did not intend the absence ofworkplace-specific exposuremeasurements automatically to triggerselection of the most protectiverespirator; instead, the Agency intendsemployers to use such equipment whenthey do not have confidence that a lessprotective respirator is sufficient. Anexample of the kind of situation thatshould trigger the use of the mostprotective respirator was provided by arepresentative of CMA, who testified(Tr. at 1707) that, when a maintenanceperson opens a closed cyclemanufacturing process to work on it forthe first time, ‘‘we don’t know what theair concentration is so we put people insupplied-air respiratory protectionunder those circumstances.’’ That is, thecompany in this case assumes thatexposures will be extremely high andselects a respirator accordingly. OSHAbelieves that the language used inparagraph (d)(1)(iii) of the final rulemakes OSHA’s intent clear, i.e., thatwhen reliable data or reasonableestimates of exposure are not available,the atmosphere must be consideredIDLH.

Finally, a few participants suggestedthat exposure estimates should only bemade by credentialed individuals (See,e.g., Ex. 54–327). OSHA agrees thatpersons trained and experienced inevaluating the respiratory hazards posedby workplace atmospheres are the mostcompetent to evaluate exposure levels,especially in the absence of currentexposure measurements. ANSI definesan ‘‘occupational health professional’’as ‘‘(a)n individual whom, byexperience and education, is competentat recognizing, evaluating, andcontrolling health hazards in theworkplace’’ (ANSI Z88.2–1992, clause3.39). This is the person who isresponsible for performing expert

evaluations under ANSI’s recommendedstandard. OSHA believes that thisdefinition has merit, and that employerswhose workplaces have highly toxicrespiratory hazards, or many differenthazardous chemicals or mixtures, aswell as other employers with theresources to do so, should utilize suchprofessionals wherever possible.However, OSHA is not specificallyincluding this requirement in the finalrule because reasonable estimations canbe conducted in many workplaces bypersons with the qualifications requiredin the final rule for the respiratoryprotection program administrator.

Paragraph (d)(1)(iv) requires that theemployer choose respirators from asufficient number of respirator modelsand sizes so that the respirator isacceptable to and correctly fits thewearer. The 1992 ANSI standardincludes a similar requirement aimed atachieving satisfactory fit and weareracceptance (Z88.2–1992, clause 9.3.1.and 9.3.2.). This provision of the finalstandard revises the correspondingproposed provision, which would haverequired employers to provide for fittesting an array of three sizes and twobrands of respirators with elastomericfacepieces. The dual intent of thisprovision was to assure that weareracceptability plays a role in respiratorselection, and that the respiratorschosen maintain their fit over the periodof use.

OSHA continues to believe that thesegoals for respirator selection areappropriate. However, OSHA waspersuaded by this record that specifyingthe number of sizes, models and brandsthat an employer must provide isunnecessary. Therefore, the finalprovision deletes the specificationlanguage for the number of sizes,models and brands that must constitutethe selection pool. Since this provisionof the final standard applies to allrespirators, the proposal’s applicationonly to ‘‘elastomeric’’ facepieces hasbeen dropped.

Most participants (Exs. 54–1, 54–5,54–75, 54–80, 54–91, 54–161, 54–208,54–214, 54–237, 54–238, 54–246, 54–263, 54–273, 54–280, 54–291, 54–287,54–350, 54–363, 54–389) endorsed theinclusion in the final rule of aperformance-based provision addressingthe selection of comfortably fittingrespirators. Thus, most comment on thisissue recognized that a sufficientassortment of respirators must beprovided so that employees will obtainacceptable fits, but that more flexibilityshould be provided in the final rule.Commenters also stated that, in somecases, a single manufacturer has avariety of respirator models sufficient to

1152 - occupational safety and health administration

Documents